2. chemistry and analytical methods 2.1 … · 2. chemistry and analytical methods ... chemistry...

20

2. CHEMISTRY AND ANALYTICAL METHODS

2.1 Physical and chemical properties

Carbon monoxide (CO) is a tasteless, odourless, colourless, non-corrosive and quite stable diatomic molecule that exists as a gas in theEarth’s atmosphere. Radiation in the visible and near-ultraviolet (UV)regions of the electromagnetic spectrum is not absorbed by carbonmonoxide, although the molecule does have weak absorption bandsbetween 125 and 155 nm. Carbon monoxide absorbs radiation in theinfrared region corresponding to the vibrational excitation of its elec-tronic ground state. It has a low electric dipole moment (0.10 debye),short interatomic distance (0.123 nm) and high heat of formationfrom atoms or bond strength (2072 kJ/mol). These observationssuggest that the molecule is a resonance hybrid of three structures(Perry et al., 1977), all of which contribute nearly equally to the nor-mal ground state. General physical properties of carbon monoxide aregiven in Table 1.

2.2 Methods for measuring carbon monoxide in ambientair

2.2.1 Introduction

Because of the low levels of carbon monoxide in ambient air,methods for its measurement require skilled personnel and sophisti-cated analytical equipment. The principles of the methodology havebeen described by Smith & Nelson (1973). A sample introductionsystem is used, consisting of a sampling probe, an intake manifold,tubing and air movers. This system is needed to collect the air samplefrom the atmosphere and to transport it to the analyser withoutaltering the original concentration. It may also be used to introduceknown gas concentrations to periodically check the reliability of theanalyser output. Construction materials for the sampling probe, intakemanifold and tubing should be tested to demonstrate that the testatmosphere composition or concentration is not altered significantly.The sample introduction system should be constructed so that itpresents no pressure drop to the analyser. At low flow and lowconcentrations, such operation may require validation.

Chemistry and Analytical Methods

21

Table 1. Physical properties of carbon monoxidea

Property Value

Molecular weight 28.01

Critical point !140 °C at 3495.7 kPa

Melting point !199 °C

Boiling point !191.5 °C

Densityat 0 °C, 101.3 kPaat 25 °C, 101.3 kPa

1.250 g/litre1.145 g/litre

Specific gravity relative to air 0.967

Solubility in waterb

at 0 °Cat 20 °Cat 25 °C

3.54 ml/100 ml (44.3 ppmm)c

2.32 ml/100 ml (29.0 ppmm)c

2.14 ml/100 ml (26.8 ppmm)c

Explosive limits in air 12.5–74.2%

Fundamental vibration transition 2143.3 cm–1

Conversion factorsat 0 °C, 101.3 kPa

at 25 °C, 101.3 kPa

1 mg/m3 = 0.800 ppmd

1 ppm = 1.250 mg/m3

1 mg/m3 = 0.873 ppmd

1 ppm = 1.145 mg/m3

a From NRC (1977).b Volume of carbon monoxide is at 0 °C, 1 atm (atmospheric pressure at sea level =

101.3 kPa).c Parts per million by mass (ppmm = :g/g).d Parts per million by volume (ppm = mg/litre).

The analyser system consists of the analyser itself and anysample preconditioning components that may be necessary. Samplepreconditioning might require a moisture control system to helpminimize the false-positive response of the analyser (e.g., the non-dispersive infrared [NDIR] analyser) to water vapour and a particulatefilter to help protect the analyser from clogging and possible chemicalinterference due to particulate buildup in the sample lines or analyserinlet. The sample preconditioning system may also include a flowmetering and flow control device to control the sampling rate to theanalyser.

A data recording system is needed to record the output of theanalyser.

EHC 213: Carbon Monoxide

22

2.2.2 Methods

A reference method or equivalent method for air qualitymeasurements is required for acceptance of measurement data. Anequivalent method for monitoring carbon monoxide can be sodesignated when the method is shown to produce results equivalentto those from the approved reference monitoring method based onabsorption of infrared radiation from a non-dispersed beam.

The designated reference methods are automated, continuousmethods utilizing the NDIR technique, which is generally accepted asbeing the most reliable method for the measurement of carbon monox-ide in ambient air. As of January 1988, no equivalent methods thatuse a principle other than NDIR have been designated for measuringcarbon monoxide in ambient air.

There have been several excellent reviews on the measurementof carbon monoxide in the atmosphere (National Air PollutionControl Administration, 1970; Driscoll & Berger, 1971; Leithe, 1971;American Industrial Hygiene Association, 1972; NIOSH, 1972;Verdin, 1973; Stevens & Herget, 1974; Harrison, 1975;Schnakenberg, 1976; NRC, 1977; Repp, 1977; Lodge, 1989; OSHA,1991a; ASTM, 1995; ISO, 1996).

2.2.2.1 Non-dispersive infrared photometry method

Currently, the most commonly used measurement technique isthe type of NDIR method referred to as gas filter correlation (Actonet al., 1973; Burch & Gryvnak, 1974; Ward & Zwick, 1975; Burch etal., 1976; Goldstein et al., 1976; Gryvnak & Burch, 1976a,b; Hergetet al., 1976; Bartle & Hall, 1977; Chaney & McClenny, 1977).

Carbon monoxide has a characteristic infrared absorption near4.6 :m. The absorption of infrared radiation by the carbon monoxidemolecule can therefore be used to measure the concentration of carbonmonoxide in the presence of other gases. The NDIR method is basedon this principle (Feldstein, 1967).

Most commercially available NDIR analysers incorporate a gasfilter to minimize interferences from other gases. They operate atatmospheric pressure, and the most sensitive analysers are able todetect minimum carbon monoxide concentrations of about 0.05 mg/m3


23

(0.044 ppm). Interferences from carbon dioxide and water vapour canbe dealt with so as not to affect the data quality. NDIR analysers withdetectors as designed by Luft (1962) are relatively insensitive to flowrate, require no wet chemicals, are sensitive over wide concentrationranges and have short response times. NDIR analysers of the newergas filter correlation type have overcome zero and span problems andminor problems due to vibrations.

2.2.2.2 Gas chromatography method

A more sensitive method for measuring low background levelsof carbon monoxide is gas chromatography (Porter & Volman, 1962;Feldstein, 1967; Swinnerton et al., 1968; Bruner et al., 1973; Dagnallet al., 1973; Tesarik & Krejci, 1974; Bergman et al., 1975; Smith etal., 1975; ISO, 1989). This technique is an automated, semi-continuous method in which carbon monoxide is separated fromwater, carbon dioxide and hydrocarbons other than methane by astripper column. Carbon monoxide and methane are then separatedon an analytical column, and the carbon monoxide is passed througha catalytic reduction tube, where it is converted to methane. Thecarbon monoxide (converted to methane) passes through a flameionization detector, and the resulting signal is proportional to theconcentration of carbon monoxide in the air. This method has beenused throughout the world. It has no known interferences and can beused to measure levels from 0.03 to 50 mg/m3 (0.026 to 43.7 ppm).These analysers are expensive and require continuous attendance bya highly trained operator to produce valid results. For high levels, auseful technique is catalytic oxidation of the carbon monoxide byHopcalite or other catalysts (Stetter & Blurton, 1976), either withtemperature-rise sensors (Naumann, 1975; Schnakenberg, 1976;Benzie et al., 1977) or with electrochemical sensors (Bay et al., 1972,1974; Bergman et al., 1975; Dempsey et al., 1975; Schnakenberg,1975; Repp, 1977). Numerous other analytical schemes have beenused to measure carbon monoxide in air.

2.2.2.3 Other analysers

Other systems to measure carbon monoxide in ambient airinclude gas chromatography/flame ionization, in which carbon mon-oxide is separated from other trace gases by gas chromatography andcatalytically converted to methane prior to detection; controlled-potential electrochemical analysis, in which carbon monoxide is


24

measured by means of the current produced in aqueous solution by itselectro-oxidation by an electro-catalytically active noble metal (theconcentration of carbon monoxide reaching the electrode is controlledby its rate of diffusion through a membrane, which depends on itsconcentration in the sampled atmosphere; Bay et al., 1972, 1974);galvanic cells that can be used to measure atmospheric carbonmonoxide continuously, in the manner described by Hersch (1964,1966); coulometric analysis, which employs a modified Hersch-typecell; mercury replacement, in which mercury vapour formed by thereduction of mercuric oxide by carbon monoxide is detectedphotometrically by its absorption of UV light at 253.7 nm; dual-isotope fluorescence, which utilizes the slight difference in theinfrared spectra of isotopes of carbon monoxide; catalyticcombustion/thermal detection, which is based on measuring thetemperature rise resulting from catalytic oxidation of the carbonmonoxide in the sample air; second-derivative spectrometry, whichutilizes a second-derivative spectrometer to process the transmissionversus wavelength function of an ordinary spectrometer to produce anoutput signal proportional to the second derivative of this function;and Fourier-transform spectroscopy, which is an extremely powerfulinfrared spectroscopic technique.

Intermittent samples may be collected in the field and lateranalysed in the laboratory by the continuous analysing techniquesdescribed above. Sample containers may be rigid (glass cylinders orstainless steel tanks) or non-rigid (plastic bags). Because of locationand cost, intermittent sampling may at times be the only practicalmethod for air monitoring. Samples can be taken over a few minutesor accumulated intermittently to obtain, after analysis, either “spot”or “integrated” results.

Additional techniques for analysing intermittent samples includecolorimetric analysis, in which carbon monoxide reacts in an alkalinesolution with the silver salt of p-sulfamoyl-benzoate to form acoloured silver sol; a National Institute of Standards and Technologycolorimetric indicating gel (incorporating palladium and molybdenumsalts), which involves colorimetric comparison with freshly preparedindicating gels exposed to known concentrations of carbon monoxide;a length-of-stain indicator method, which uses an indicator tubecontaining potassium palladosulfite; and frontal analysis, in which airis passed over an adsorbent until equilibrium is established between


25

the concentration of carbon monoxide in the air and the concentrationof carbon monoxide on the adsorbent.

A simple and inexpensive measurement technique uses detectortubes (indicator tubes) (Leichnitz, 1993). This method is widelyapplied in industrial hygiene and is suitable for analysis of highlypolluted atmospheric air. The measurement with Dräger tubes(Drägerwerk, 1994) is based on the reaction: 5CO + I2O5 ÷ I2 +5CO2. The iodine-coloured layer in the tube corresponds in length tothe carbon monoxide concentration in the sample.

2.2.3 Measurement using personal monitors

Until the 1960s, most of the data available on ambient carbonmonoxide concentrations came from fixed monitoring stations oper-ated routinely in urban areas. The accepted measurement techniquewas NDIR spectrometry, but the instruments were large and cumber-some, often requiring vibration-free, air-conditioned enclosures.Without a portable, convenient monitor for carbon monoxide, it wasextremely difficult to measure carbon monoxide concentrations accu-rately in the microenvironments that people usually visited.

Ultimately, small personal exposure monitors (PEMs) weredeveloped that could measure carbon monoxide concentrations con-tinuously over time and store the readings automatically on internaldigital memories (Ott et al., 1986). These small PEMs made possiblethe large-scale field studies on human exposure to carbon monoxidein Denver, Colorado, and Washington, DC, USA, in the winter of1982–83 (Akland et al., 1985).

2.2.4 Carbon monoxide detectors/alarms

Carbon monoxide detectors have been designed like residentialsmoke detectors — to be low cost, yet provide protection from acatastrophic event by sounding an audible alarm. The carbon mon-oxide detector industry is young, however, and is in a stage of rapidgrowth. In the USA, an estimated 7–8 million detectors have beenpurchased since the early 1990s, but the numbers used in homes willcontinue to rise as local municipalities change building codes torequire the installation of carbon monoxide detectors in new residen-tial structures containing combustion-source appliances, stoves orfireplaces.


26

Currently available carbon monoxide detectors are based on aninteractive-type sensor (e.g., tin oxide, Figaro-type gel cell, fuel cell,artificial haemoglobin) that relies on direct interaction betweencarbon monoxide and the sensitive element in order to generate aresponse. They are battery-powered, alternating current-powered orboth. The most popular alternating current-powered detectors have aheated metallic sensor that reacts with carbon monoxide; the battery-powered detectors have a chemically treated gel disk that darkens withexposure to carbon monoxide or a fuel cell. Small, inexpensive carbonmonoxide detection cards or tablets that require frequent visualinspection of colour changes do not sound an alarm and are notrecommended as a primary detector.

Carbon monoxide detectors are sensitive to location and environ-mental conditions, including temperature, relative humidity and thepresence of other interfering gases. They may also become less stablewith time. For example, they should not be installed in dead-space air(i.e., near ceilings), near windows or near doors where there is a lotof air movement, and they should not be exposed to temperature orhumidity extremes. Excessive heat or cold will affect performance,and humidity extremes will affect the activation time. Utilization ofnon-interactive infrared technology (e.g., NDIR) in indoor carbonmonoxide detection would overcome all of the shortcomings of thecurrently available carbon monoxide detectors.

In the USA, a new voluntary standard for carbon monoxidedetectors was published in 1992 by the Underwriters Laboratories (ULStandard 2034) and revised in 1995. This standard provides alarmrequirements for detectors that are based on both the carbon monoxideconcentration and the exposure time. It is designed so that an alarmis activated within 90 min of exposure to 110 mg/m3 (100 ppm),within 35 min of exposure to 230 mg/m3 (200 ppm) or within 15 minof exposure to 460 mg/m3 (400 ppm) (i.e., when exposures areequivalent to 10% carboxyhaemoglobin [COHb]; see section 2.3).Approximately 15 manufacturers produce detectors listed under ULStandard 2034.

Because UL Standard 2034 covers a wide range of exposureconditions, there has been some ambiguity about its interpretation.For example, it is not clear if a detector meets the standard if thealarm is activated anytime between 5 and 90 min in the presence of110 mg carbon monoxide/mg3 (100 ppm). In fact, alarm sensitivities


27

are still a problem for the industry, and further discussion anddirection are needed. Moreover, the 10% carboxyhaemoglobin levelis protective of healthy individuals only (see chapter 8). It would benecessary to avoid exposures to 10 mg carbon monoxide/m3 (9 ppm)for 8 h or 29 mg/m3 (25 ppm) for 1 h in order to protect sensitiveindividuals with coronary heart disease at the 3%carboxyhaemoglobin level. Thus, current detectors provide warningagainst carbon monoxide levels that are protective of the healthypopulation only. Despite these limitations, carbon monoxide detectorsare reliable and effective, continue to improve and should berecommended for use in homes in addition to smoke detectors and firealarms.

2.3 Biological monitoring

A unique feature of carbon monoxide exposure is that there is abiological marker of the dose that the individual has received: thelevel of carbon monoxide in the blood. This level may be calculatedby measuring carboxyhaemoglobin in blood or carbon monoxide inexhaled breath.

2.3.1 Blood carboxyhaemoglobin measurement

The level of dissolved carbon monoxide in blood is normallybelow the level of detection but may be of importance in thetransportation of carbon monoxide between cells and tissues (seechapter 6). Thus, the blood level of carbon monoxide isconventionally represented as a percentage of the total haemoglobinavailable (i.e., the percentage of haemoglobin that is in the form ofcarboxyhaemoglobin, or simply percent carboxyhaemoglobin).

Any technique for the measurement of carboxyhaemoglobin inblood must be specific and must have sufficient sensitivity andaccuracy for the purpose of the values obtained. The majority oftechnical methods that have been published on measurement ofcarbon monoxide in blood have been for forensic purposes. Thesemethods are less accurate than generally required for themeasurement of low levels of carboxyhaemoglobin (<5%). Bloodlevels of carbon monoxide resulting from exposure to existing ambientlevels of carbon monoxide would not be expected to exceed 5%carboxyhaemoglobin in non-smoking subjects. The focus of theforensic methods has been the reliability of measurements over the


28

entire range of possible values: from less than 1% to 100%carboxyhaemoglobin. These forensically oriented methods areadequate for the intended use of the values and the non-ideal storageconditions of the samples being analysed.

In the areas of exposure assessment and low-level health effectsof carbon monoxide, it is more important to know the accuracy of anymethod in the low-level range of <5% carboxyhaemoglobin. There islittle agreement upon acceptable reference methods in this range, norare there accurate reference standards available in this range. The useof techniques that have unsubstantiated accuracy in the low range ofcarboxyhaemoglobin levels can lead to considerable differences inestimations of exposure conditions. Measurement of low levels ofcarboxyhaemoglobin demands careful evaluation because of the impli-cations, based upon these data, for the setting of air quality standards.Therefore, this section will focus on the methods that have beenevaluated at levels below 10% carboxyhaemoglobin and the methodsthat have been extensively used in assessing exposure to carbonmonoxide.

The measurement of carbon monoxide in blood can be accom-plished by a variety of techniques, both destructive and non-destructive. Carboxyhaemoglobin can be determined non-destructivelyby observing the change in the absorption spectrum in either the Soretor visible region brought about by the combination of carbon monox-ide with haemoglobin. With present optical sensing techniques,however, all optical methods are limited in sensitivity to approxi-mately 1% of the range of expected values. If attempts are made toexpand the lower range of absorbances, sensitivity is lost on the upperend where, in the case of carboxyhaemoglobin, total haemoglobin ismeasured. For example, in the spectrophotometric method describedby Small et al. (1971), a change in absorbance equal to the limit ofresolution of 0.01 units can result in a difference in 0.6% carboxy-haemoglobin. Therefore, optical techniques cannot be expected toobtain the resolution that is possible with other means of detection ofcarbon monoxide (Table 2).

The more sensitive (higher-resolution) techniques require therelease of the carbon monoxide from the haemoglobin into a gasphase; the carbon monoxide can then be detected directly by (1) infra-red absorption (Maas et al., 1970) following separation using


29

Table 2. Representative methods for the analysis of carbon monoxide in blooda

Source Method Resolutionb

(ml/dl)CV (%)c Reference

methodrd

Gasometric detection

Scholander & Roughton(1943)

Syringecapillary

0.02 2–4 Van Slyke NDe

Horvath & Roughton(1942)

Van Slyke 0.03 6 Van Slyke–Neill

ND

Spectrophotometric detection

Coburn et al. (1964) Infrared 0.006 1.8 Van Slyke–Syringe

ND

Small et al. (1971) Spectro-photometry

0.12 ND Flameionization

ND

Maas et al. (1970) CO-Oximeter(IL-182)

0.21 5 Spectro-photometric

ND

Brown (1980) CO-Oximeter(IL-282)

0.2 5 Flameionization

0.999

Gas chromatography

Ayres et al. (1966) Thermalconductivity

0.001 2 ND ND

Goldbaum et al. (1986) Thermalconductivity

ND 1.35 Flameionization

0.996

McCredie & Jose (1967) Thermalconductivity

0.005 1.8 ND ND

Dahms & Horvath (1974) Thermalconductivity

0.006 1.7 Van Slyke 0.983

Collison et al. (1968) Flameionization

0.002 1.8 Van Slyke ND

Kane (1985) Flameionization

ND 6.2 CO-Oximeter

1.00

Vreman et al. (1984) Mercuryvapour

0.002 2.2 ND ND

a Modified from US EPA (1991d).b The resolution is the smallest detectable amount of carbon monoxide or the smallest

detectable difference between samples. c Coefficient of variation (CV) was computed on samples containing less than 15%

carboxyhaemoglobin, where possible.d The r value is the correlation coefficient between the technique reported and the

reference method used to verify its accuracy. e ND indicates that no data were available.


30

gas chromatography, (2) the difference in thermal conductivitybetween carbon monoxide and the carrier gas (Ayres et al., 1966;McCredie & Jose, 1967; Dahms & Horvath, 1974; Goldbaum et al.,1986; Horvath et al., 1988b; Allred et al., 1989b), (3) the amount ofionization following quantitative conversion of carbon monoxide tomethane and ionization of the methane (Collison et al., 1968; Dennis& Valeri, 1980; Guillot et al., 1981; Clerbaux et al., 1984; Kane,1985; Katsumata et al., 1985; Costantino et al., 1986) or (4) therelease of mercury vapour resulting from the combination of carbonmonoxide with mercuric oxide (Vreman et al., 1984).

2.3.1.1 Sample handling

Carbon monoxide bound to haemoglobin is a relatively stablecompound that can be dissociated by exposure to oxygen or UVradiation (Horvath & Roughton, 1942; Chace et al., 1986). If theblood sample is maintained in the dark under cool, sterile conditions,the carbon monoxide content will remain stable for a long period oftime. Various investigators have reported no decrease in percentcarboxyhaemoglobin over 10 days (Collison et al., 1968), 3 weeks(Dahms & Horvath, 1974), 4 months (Ocak et al., 1985) and6 months (Vreman et al., 1984). The blood collection system used caninfluence the carbon monoxide level, because some ethylenediamine-tetraacetic acid vacutainer tube stoppers contain carbon monoxide(Vreman et al., 1984). The stability of the carbon monoxide contentin properly stored samples does not indicate that constant values willbe obtained by all techniques of analysis. The spectrophotometricmethods are particularly susceptible to changes in optical qualities ofthe sample, resulting in small changes in carboxyhaemoglobin withstorage (Allred et al., 1989b).

Therefore, the care needed to make a carboxyhaemoglobin deter-mination depends upon the technique that is being utilized. It appearsas though measurement of low levels of carboxyhaemoglobin withoptical techniques should be conducted as soon as possible followingcollection of the samples.

2.3.1.2 Potential reference methods

Exposure to carbon monoxide at equilibrium conditions resultsin carboxyhaemoglobin levels of between 0.1 and 0.2% for eachmilligram of carbon monoxide per cubic metre air (part per million).


31

A reference technique for the measurement of carboxyhaemoglobinshould be able to discriminate between two blood samples with adifference of 0.1% carboxyhaemoglobin (approximately 0.02 ml/dl).To accomplish this task, the coefficient of variation (standard devia-tion of repeated measures on any given sample divided by the meanof the values times 100) of the method should be less than 5%, so thatthe two values that are different by 0.1 percentage points can be statis-tically proven to be distinct. In practical terms, a reference methodshould have the sensitivity to detect approximately 0.025% carboxy-haemoglobin to provide this level of confidence in the valuesobtained.

The accurate measurement of carboxyhaemoglobin requires thequantitation of the content of carbon monoxide released from haemo-globin in the blood. Optically based techniques have limitations ofresolution and specificity due to the potential interference from manysources. The techniques that can be used as reference methods involvethe quantitative release of carbon monoxide from the haemoglobinfollowed by the measurement of the amount of carbon monoxidereleased. Classically, this quantitation was measured manometricallywith a Van Slyke apparatus (Horvath & Roughton, 1942) or aRoughton-Scholander syringe (Roughton & Root, 1945). These tech-niques have served as the “Gold Standard” in this field for almost50 years. However, there are limitations of resolution with thesetechniques at the lower ranges of carboxyhaemoglobin. The gasomet-ric standard methodology has been replaced with headspace extractionfollowed by the use of solid-phase gas chromatographic separationwith several different types of detection: thermal conductivity, flameionization and mercury vapour reduction. The carbon monoxide in theheadspace can also be quantitated by infrared detection, which can becalibrated with gas standards. However, there is no general agreementthat any of the more sensitive methods of carbon monoxide analysisare acceptable reference methods.

The following techniques all conform to all the requirements ofa reference method:

(1) Flame ionization detection: This technique requires theseparation of carbon monoxide from the other headspacegases and the reduction of the carbon monoxide to methaneby catalytic reduction.


32

(2) Thermal conductivity detection: This technique usesvacuum extraction of carbon monoxide from blood in a VanSlyke apparatus and gas chromatographic separation withthermal conductivity analysis of the carbon monoxide.

(3) Infrared detection: This technique uses a method forextracting carbon monoxide from blood under normalatmospheric conditions and then injecting the headspacegas into an infrared analyser.

The conventional means of representing the quantity of carbonmonoxide in a blood sample is the percent carboxyhaemoglobin: thepercentage of the total carbon monoxide combining capacity that is inthe form of carboxyhaemoglobin. This is conventionally determinedby the use of the following formula:

% COHb = [CO content/(Hb × 1.389)] × 100 (2-1)

where CO content is the carbon monoxide concentration, measuredin millilitres per decilitre blood at standard temperature and pressure,dry; Hb is the haemoglobin concentration, measured in grams perdecilitre blood; and 1.389 is the stoichiometric combining capacity ofcarbon monoxide for haemoglobin in units of millilitres of carbonmonoxide per gram of haemoglobin at standard temperature and pres-sure, dry; however, in practice, a value of 1.36 ml carbon monoxide/ghaemoglobin is used for the oxygen capacity of normal human bloodbecause it is impossible to achieve 100% haemoglobin saturation.

The analytical methods that quantify the carbon monoxidecontent in blood require the conversion of these quantities to percentcarboxyhaemoglobin. The product of the haemoglobin and thetheoretical combining capacity (1.389, according to InternationalCommittee for Standardization in Haematology, 1978) yields thecarbon monoxide capacity. With the use of capacity and the measuredcontent, the percentage of carbon monoxide capacity (percentcarboxyhaemoglobin) is calculated. To be absolutely certain of theaccuracy of the haemoglobin measurement, the theoretical valueshould be routinely substantiated by direct measurement (internalvalidation) of the haemoglobin–carbon monoxide combining capacity.The total carbon monoxide–haemoglobin combining capacity shouldbe determined as accurately as the content of carbon monoxide. Theerror of the techniques that measure carbon monoxide content is


33

dependent on the error in haemoglobin analysis for the final form ofthe data, percent carboxyhaemoglobin. Therefore, the actual carbonmonoxide–haemoglobin combining capacity should be measured andcompared with the calculated value based upon the reference methodfor haemoglobin measurement. The measurement of carbon monox-ide–haemoglobin combining capacity can be routinely performed byequilibration of a blood sample with carbon monoxide (Allred et al.,1989b).

The standard methods for haemoglobin determination involve theconversion of all species of haemoglobin to cyanomethaemoglobinwith the use of a mixture of potassium ferricyanide, potassium cyanideand sodium bicarbonate.

2.3.1.3 Other methods of measurement

There is a wide variety of other techniques that have beendescribed for the analysis of carbon monoxide in blood. These meth-ods include UV-visible spectrophotometry (Small et al., 1971; Brown,1980; Zwart et al., 1984, 1986), magnetic circular dichroism spectros-copy (Wigfield et al., 1981), photochemistry (Sawicki & Gibson,1979), gasometric methods (Horvath & Roughton, 1942; Roughton &Root, 1945) and a calorimetric method (Sjostrand, 1948a). Not all ofthese methods have been as well characterized for the measurementof low levels of carboxyhaemoglobin as those listed above as potentialreference methods.

1) Spectrophotometric methods

The majority of the techniques are based upon optical detectionof carboxyhaemoglobin, which is more rapid than the referencetechniques because it does not involve extraction of the carbon mon-oxide from the blood sample. These direct measurements also enablethe simultaneous measurement of several species of haemoglobin,including reduced haemoglobin, oxyhaemoglobin (O2Hb) and car-boxyhaemoglobin. The limitations of the spectrophotometric tech-niques have been reviewed by Kane (1985). The optical methodsutilizing UV wavelengths require dilution of the blood sample, whichcan lead to the loss of carbon monoxide as a result of competition withthe dissolved oxygen in solution. Removing the dissolved oxygen withdithionite can lead to the formation of sulfhaemoglobin, which inter-feres with the measurement of carboxyhaemoglobin (Rai & Minty,


34

1987). Another limitation is that the absorption maxima (and spectralcurves) are not precisely consistent between individuals. This may bedue to slight variations in types of haemoglobin in subjects. For thesereasons, the techniques using fixed-wavelength measurement pointswould not be expected to be as precise, accurate or specific as theproposed reference methods mentioned above.

2) CO-Oximeter measurements of carboxyhaemoglobin

The speed of measurement and relative accuracy of spectrophoto-metric measurements over the entire range of expected values led tothe development of CO-Oximeters. These instruments utilize fromtwo to seven wavelengths in the visible region for the determinationof proportions of oxyhaemoglobin, carboxyhaemoglobin, reduced hae-moglobin and methaemoglobin. The proportion of each species ofhaemoglobin is determined from the absorbance and molar extinctioncoefficients at present wavelengths. All of the commercially availableinstruments provide rapid results for all the species of haemoglobinbeing measured. In general, the manufacturers’ listed limit ofaccuracy for all of the instruments is 1% carboxyhaemoglobin.However, this level of accuracy is not suitable for measurementsassociated with background carbon monoxide levels (<2%carboxyhaemoglobin) because it corresponds to errors exceeding 50%.The precision of measurement for these instruments is excellent andhas misled users regarding the accuracy of the instruments. Therelatively modest level of accuracy is adequate for the design purposesof the instruments; however, at low levels of carboxyhaemoglobin, theability of the instruments to measure the percent carboxyhaemoglobinaccurately is limited.

2.3.2 Carbon monoxide in expired breath

Carbon monoxide levels in expired breath can be used to estimatethe levels of carbon monoxide in the subject’s blood. The basicdeterminants of carbon monoxide levels in alveolar air have beendescribed by Douglas et al. (1912), indicating that there arepredictable equilibrium conditions that exist between carbonmonoxide bound to haemoglobin and the partial pressure of thecarbon monoxide in the blood. The equilibrium relationship forcarbon monoxide between blood and the gas phase to which the bloodis exposed can be described as follows:


35

PCO/PO2 = M (% COHb/% O2Hb) (2-2)

where PCO is the partial pressure of carbon monoxide in the blood, PO2

is the partial pressure of oxygen in the blood, M is the Haldanecoefficient (reflecting the relative affinity of haemoglobin for oxygenand carbon monoxide), % COHb is the percentage of total haemo-globin combining capacity bound with carbon monoxide, and % O2Hbis the percentage of total haemoglobin combining capacity bound withoxygen.

The partial pressure of carbon monoxide in the arterial blood willreach a steady-state value relative to the partial pressure of carbonmonoxide in the alveolar gas. Therefore, by measuring theend-expired breath from a subject’s lungs, one can measure theend-expired carbon monoxide partial pressure and, with the use of theHaldane relationship, estimate the blood level of carboxyhaemoglobin.This measurement will always be an estimate, because the Haldanerelationship is based upon attainment of an equilibrium, which doesnot occur under physiological conditions.

The measurement of carbon monoxide levels in expired breathto estimate blood levels is based upon application of the Haldanerelationship to gas transfer in the lung (Eq. 2-2). For example, whenthe oxygen partial pressure is increased in the alveolar gas, it ispossible to predict the extent to which the partial pressure of carbonmonoxide will increase in the alveolar gas. This approach is limited,however, because of the uncertainty associated with variables that areknown to influence gas transfer in the lung and that mediate the directrelationship between liquid-phase gas partial pressures and air-phasepartial pressures.

The basic mechanisms that are known to influence carbonmonoxide transfer in the lung have been identified through theestablishment of techniques to measure pulmonary diffusion capacityfor carbon monoxide. Some of the factors that can result in decreaseddiffusion capacity for carbon monoxide (altering the relationshipbetween expired carbon monoxide pressures and carboxyhaemoglobinlevels) are increased membrane resistance, intravascular resistance,age, alveolar volume, pulmonary vascular blood volume, pulmonaryblood flow and ventilation/perfusion inequality (Forster, 1964). Theextent to which each of these variables actually contributes to the


36

variability in the relationship has not been experimentallydemonstrated. There are few experiments that focus on the factorsleading to variability in the relationship between alveolar carbonmonoxide and percent carboxyhaemoglobin at the levels ofcarboxyhaemoglobin currently deemed to be of regulatory importance.This may be due in part to the difficulties in working with analyticaltechniques, particularly blood techniques, that are very close to theirlimits of reproducibility.

The expired breath method for obtaining estimates of blood levelsof carbon monoxide has a distinct advantage for monitoring largenumbers of subjects, because of the non-invasive nature of themethod. Other advantages include the ability to obtain an instan-taneous reading and the ability to take an immediate replicate samplefor internal standardization. The breath-holding technique forenhancing the normal carbon monoxide concentration in exhaledbreath has been widely used; however, it should be noted that theabsolute relationship between breath-hold carbon monoxide pressuresand blood carbon monoxide pressures has not been thoroughlyestablished for carboxyhaemoglobin levels below 5%. The breath-holding method allows time (20 s) for diffusion of carbon monoxideinto the alveolar air so that carbon monoxide levels are higher thanlevels following normal tidal breathing.

Partial pressures of carbon monoxide in expired breath are highlycorrelated with percent carboxyhaemoglobin levels over a wide rangeof carboxyhaemoglobin levels. The accuracy of the breath-holdmethod is unknown owing to the lack of paired sample analyses ofcarbon monoxide partial pressures in exhaled breath and concurrentcarboxyhaemoglobin levels in blood utilizing a sensitive referencemethod. No one has attempted to determine the error of estimateinvolved in applying group average regression relationships to theaccurate determination of carboxyhaemoglobin. Therefore, the extrap-olation of breath-hold carbon monoxide partial pressures to actualcarboxyhaemoglobin levels must be made with reservation until theaccuracy of this method is better understood.

2.3.2.1 Measurement methods

Ventilation in healthy individuals involves air movement throughareas in the pulmonary system that are primarily involved in eitherconduction of gas or gas exchange in the alveoli. In a normal breath(tidal volume), the proportion of the volume in the non-gas exchang-


37

ing area is termed the dead space. In the measurement of carbonmonoxide in the exhaled air, the dead-space gas volume serves todilute the alveolar carbon monoxide concentration. Several methodshave been developed to account for the dead-space dilution. Theseinclude the mixed expired gas technique, which uses the Bohrequation to determine the physiological dead space; the breath-holdtechnique, a method of inspiration to total lung capacity followed bya breath-hold period of various durations (a breath-hold time of 20 swas found to provide near-maximal values for carbon monoxidepressures); and the rebreathing technique, in which 5 litres of oxygenare rebreathed for 2–3 min while the carbon dioxide is removed.

Kirkham et al. (1988) compared all three techniques for measur-ing expired carbon monoxide to predict percent carboxyhaemoglobin.The rebreathing and breath-hold methods both yield approximately20% higher levels of “alveolar” carbon monoxide than does the Bohrcomputation from mixed expired gas. Both the mixed expired andbreath-holding techniques show a significant decline in the alveolarcarbon monoxide tension when the subject is standing. Therefore,measurements of expired carbon monoxide must be made in the samebody position relative to control measurements or reference measure-ments.

2.3.2.2 Potential limitations

The measurement of exhaled breath has the advantages of ease,speed, precision and greater subject acceptance over measurement ofblood carboxyhaemoglobin. However, the accuracy of the breathmeasurement procedure and the validity of the Haldane relationshipbetween breath and blood at low environmental carbon monoxideconcentrations remain in question.

38

3. SOURCES OF CARBON MONOXIDE IN THEENVIRONMENT

3.1 Introduction

Carbon monoxide is produced by both natural and anthropogenicprocesses. About half of the carbon monoxide is released at theEarth’s surface, and the rest is produced in the atmosphere. Manypapers on the global sources of carbon monoxide have been publishedover the last 20 years; whether most of the carbon monoxide in theatmosphere is from human activities or from natural processes hasbeen debated for nearly as long.

The recent budgets that take into account previously publisheddata suggest that human activities are responsible for about 60% ofthe carbon monoxide in the non-urban troposphere, and naturalprocesses account for the remaining 40%. It also appears thatcombustion processes directly produce about 40% of the annualemissions of carbon monoxide (Jaffe, 1968, 1973; Robinson &Robbins, 1969, 1970; Swinnerton et al., 1971), and oxidation ofhydrocarbons makes up most of the remainder (about 50%) (Went,1960, 1966; Rasmussen & Went, 1965; Zimmerman et al., 1978;Hanst et al., 1980; Greenberg et al., 1985), along with other sourcessuch as the oceans (Swinnerton et al., 1969; Seiler & Junge, 1970;Lamontagne et al., 1971; Linnenbom et al., 1973; Liss & Slater, 1974;Seiler, 1974; Seiler & Schmidt, 1974; Swinnerton & Lamontagne,1974; NRC, 1977; Bauer et al., 1980; Logan et al., 1981; DeMoreet al., 1985) and vegetation (Krall & Tolbert, 1957; Wilks, 1959;Siegel et al., 1962; Seiler & Junge, 1970; Bidwell & Fraser, 1972;Seiler, 1974; NRC, 1977; Seiler & Giehl, 1977; Seiler et al., 1978;Bauer et al., 1980; Logan et al., 1981; DeMore et al., 1985). Some ofthe hydrocarbons that eventually end up as carbon monoxide are alsoproduced by combustion processes, constituting an indirect source ofcarbon monoxide from combustion. These conclusions are summar-ized in Table 3, which is adapted from the 1981 budget of Loganet al., in which most of the previous work was incorporated (Loganet al., 1981; WMO, 1986). The total emissions of carbon monoxideare about 2600 million tonnes per year. Other budgets by Volz et al.(1981) and by Seiler & Conrad (1987) have been reviewed byWarneck (1988). Global emissions between 2000 and 3000 milliontonnes per year are consistent with these budgets.

Sources of Carbon Monoxide in the Environment

39

Table 3. Sources of carbon monoxidea

Carbon monoxide production (million tonnes peryear)b

Anthropogenic Natural Global Range

Directly from combustion

Fossil fuelsForest clearingSavanna burningWood burningForest fires

50040020050—

————30

5004002005030

400–1000200–800100–40025–15010–50

Oxidation of hydrocarbons

Methanec

Non-methane hydrocarbons30090

300600

600690

400–1000300–1400

Other sources

PlantsOceans

——

10040

10040

50–20020–80

Totals (rounded) 1500 1100 2600 2000–3000

a Adapted from Logan et al. (1981) and revisions reported by the WMO (1986). b All estimates are expressed to one significant figure. The sums are rounded to two

significant digits.c Half the production of carbon monoxide from the oxidation of methane is attributed to

anthropogenic sources and the other half to natural sources based on the budget ofmethane from Khalil & Rasmussen (1984c).

3.2 Principles of formation by source category

Carbon monoxide is produced in the atmosphere by reactions ofhydroxyl radicals with methane and other hydrocarbons, both anthro-pogenic and natural, as well as by the reactions of alkenes with ozoneand of isoprene and terpenes with hydroxyl radicals and ozone.

Carbon monoxide is also produced at the Earth’s surface duringthe combustion of fuels. The burning of any carbonaceous fuel pro-duces two primary products: carbon dioxide and carbon monoxide.The production of carbon dioxide predominates when the air oroxygen supply is in excess of the stoichiometric needs for completecombustion. If burning occurs under fuel-rich conditions, with less airor oxygen than is needed, carbon monoxide will be produced inabundance. In past years, most of the carbon monoxide and carbon


40

dioxide formed were simply emitted into the atmosphere. In recentyears, concerted efforts have been made to reduce ambient air con-centrations of materials that are potentially harmful to humans. Muchcarbon monoxide, most notably from mobile sources, is converted tocarbon dioxide, which is then emitted into the atmosphere.

Emission source categories in the USA (US EPA, 1991b) aredivided into five individual categories: (1) transportation, (2) station-ary source fuel combustion, (3) industrial processes, (4) solid wastedisposal and (5) miscellaneous.

Transportation sources include emissions from all mobilesources, including highway and off-highway motor vehicles. Highwaymotor vehicles include passenger cars, trucks, buses and motorcycles.Off-highway vehicles include aircraft, locomotives, vessels andmiscellaneous engines such as farm equipment, industrial andconstruction machinery, lawnmowers and snowmobiles.

Emission estimates from gasoline- and diesel-powered motorvehicles are based upon vehicle-mile (vehicle-kilometre) tabulationsand emission factors. Eight vehicle categories are considered:(1) light-duty gasoline vehicles (mostly passenger cars), (2) light-dutydiesel passenger cars, (3) light-duty gasoline trucks (weighing lessthan 6000 lb [2.7 tonnes]), (4) light-duty gasoline trucks (weighing6000–8500 lb [2.7–3.9 tonnes]), (5) light-duty diesel trucks,(6) heavy-duty gasoline trucks and buses, (7) heavy-duty diesel trucksand buses and (8) motorcycles. The emission factors used are basedon the US Environmental Protection Agency’s (EPA) mobile sourceemission factor model, developed by the EPA Office of MobileSources, which uses the latest available data to estimate averagein-use emissions from highway vehicles.

Aircraft emissions are based on emission factors and aircraftactivity statistics reported by the Federal Aviation Administration(1988). Emissions are based on the number of landing–take-off cycles.Any emissions in cruise mode, which is defined to be above 3000 ft(1000 m), are ignored. Average emission factors for each year, whichtake into account the national mix of aircraft types for generalaviation, military and commercial aircraft, are used to compute theemissions.


41

In the USA, the Department of Energy reports consumption ofdiesel fuel and residual fuel oil by locomotives (US Department ofEnergy, 1988a). Average emission factors applicable to diesel fuelconsumption were used to calculate emissions. Vessel use of dieselfuel, residual oil and coal is also reported by the Department ofEnergy (US Department of Energy, 1988a,b). Gasoline use is basedon national boat and motor registrations, coupled with a use factor(gallons [litres] per motor per year) (Hare & Springer, 1973) andmarine gasoline sales (US Department of Transportation, 1988).Emission factors from EPA Report No. AP-42 are used to computeemissions (US EPA, 1985).

Gasoline and diesel fuel are consumed by off-highway vehiclesin substantial quantities. The fuel consumption is divided into severalcategories (e.g., farm tractors, other farm machinery, constructionequipment, industrial machinery, snowmobiles and small generalutility engines such as lawnmowers and snowblowers). Fuel use isestimated for each category from estimated equipment population andan annual use factor of gallons (litres) per unit per year (Hare &Springer, 1973), together with reported off-highway diesel fueldeliveries (US Department of Energy, 1988a) and off-highwaygasoline sales (US Department of Transportation, 1988).

Stationary combustion equipment, such as coal-, gas- or oil-firedheating or power generating plants, generates carbon monoxide as aresult of improper or inefficient operating practices or inefficientcombustion techniques. The specific emission factors for stationaryfuel combustors vary according to the type and size of the installationand the fuel used, as well as the mode of operation. The US EPA’scompilation of air pollutant emission factors provides emission dataobtained from source tests, material balance studies, engineeringestimates and so forth for the various common emission categories.For example, coal-fired electricity generating plants report coal use tothe US Department of Energy (1988b,c). Distillate oil, residual oil,kerosene and natural gas consumed by stationary combustors are alsoreported by user category to the US Department of Energy (1988a).Average emission factors from EPA Report No. AP-42 (US EPA,1985) were used to calculate the emission estimates. The consumptionof wood in residential wood stoves has likewise been estimated by theUS Department of Energy (1982, 1984).


42

In addition to fuel combustion, certain other industrial processesgenerate and emit varying quantities of carbon monoxide into the air.The lack of published national data on production, type of equipmentand controls, as well as an absence of emission factors, makes itimpossible to include estimates of emissions from all industrial proc-ess sources.

Solid waste carbon monoxide emissions result from the com-bustion of wastes in municipal and other incinerators, as well as fromthe open burning of domestic and municipal refuse.

Miscellaneous carbon monoxide emissions result from theburning of forest and agricultural materials, smouldering coal refusematerials and structural fires.

The Forest Service of the US Department of Agriculture pub-lishes information on the number of forest fires and the acreageburned (US Forest Service, 1988). Estimates of the amount of materialburned per acre are made to determine the total amount of materialburned. Similar estimates are made to account for managed burningof forest areas. Average emission factors were applied to the quantitiesof materials burned to calculate emissions.

A study was conducted by the US EPA (Yamate, 1974) to obtain,from local agricultural and pollution control agencies, estimates of thenumber of acres and estimated quantity of material burned per acre inagricultural burning operations. These data have been updated andused to estimate agricultural burning emissions, based on averageemission factors.

Estimates of the number of burning coal refuse piles existing inthe USA are made in reports by the Bureau of Mines. McNay (1971)presents a detailed discussion of the nature, origin and extent of thissource of pollution. Rough estimates of the quantity of emissions wereobtained using this information by applying average emission factorsfor coal combustion. It was assumed that the number of burning refusepiles decreased to a negligible level by 1975.

The US Department of Commerce publishes, in its statisticalabstracts, information on the number and types of structures damagedby fire (US Department of Commerce, 1987). Emissions were


43

estimated by applying average emission factors for wood combustionto these totals.

The estimated total annual carbon monoxide emissions from thevarious source categories in the USA for 1970, 1975 and 1980–1990(US EPA, 1991b) indicate that carbon monoxide emissions from allanthropogenic sources declined from 101.4 million tonnes (111.8 mil-lion short tons) in 1970 to 60.1 million tonnes (66.2 million shorttons) in 1990. The majority, about 63%, of the carbon monoxideemissions total comes from transportation sources, 12% comes fromstationary source fuel combustion, 8% comes from industrialprocesses, 3% comes from solid waste and 14% comes frommiscellaneous sources.

The single largest contributing source of carbon monoxideemissions is highway vehicles, which emitted an estimated 50% of thenational total in 1990. Because of the implementation of the FederalMotor Vehicle Control Program, carbon monoxide emissions fromhighway vehicles declined 54%, from 65.3 to 30.3 million tonnes, inthe period 1970–1990. Fig. 1 displays the trend in estimated carbonmonoxide emissions from the major highway vehicle categories from1970 to 1990. Although the total annual vehicle-miles (vehicle-kilometres) travelled continue to increase in the USA (by 37% just inthe period 1981–1990), total carbon monoxide emissions fromhighway vehicles have continued to decrease as a result of the FederalMotor Vehicle Control Program-mandated air pollution controldevices on new vehicles.

Carbon monoxide emissions from other sources have alsogenerally decreased. In 1970, emissions from burning of agriculturalcrop residues were greater than in more recent years. Solid wastedisposal emissions have also decreased as the result of implementationof regulations limiting or prohibiting burning of solid waste in manyareas. Emissions of carbon monoxide from stationary source fuelcombustion occur mainly from the residential sector. These emissionswere reduced somewhat through the mid-1970s as residential consum-ers converted to natural gas, oil or electric heating equipment. Recentgrowth in the use of residential wood stoves has reversed this trend,but increased carbon monoxide emissions from residential sourcescontinue to be small compared with highway vehicle emissions.Nevertheless, in 1990, residential wood combustion accounted for

Fig. 1. Estimated emission of carbon monoxide from gasoline-fuelled highway vehicles in the USA (adapted from US EPA, 1991a).


45

about 10% of national carbon monoxide emissions, more than anysource category except highway vehicles. Carbon monoxide emissionsfrom industrial processes have generally been declining since 1970 asthe result of the obsolescence of a few high-polluting processes suchas manufacture of carbon black by the channel process andinstallation of controls on other processes.

Considerable effort has been made to reduce emissions of carbonmonoxide and other pollutants to the atmosphere. Because the auto-mobile engine is recognized to be the major source of carbon monox-ide in most urban areas, special attention is given to the control ofautomotive emissions. Generally, the approach has beentechnological: reduction of carbon monoxide emissions to theatmosphere either by improving the efficiency of the combustionprocesses, thereby increasing the yield of carbon dioxide anddecreasing the yield of carbon monoxide, or by applying secondarycatalytic combustion reactors to the waste gas stream to convertcarbon monoxide to carbon dioxide.

The development and application of control technology to reduceemissions of carbon monoxide from combustion processes generallyhave been successful and are continuing to receive deserved attention.The reduction of carbon monoxide emissions from 7.0 to 3.4 g/mi(from 4.4 to 2.1 g/km), scheduled for the 1981 model year, wasdelayed 2 years, reflecting in part the apparent difficulty encounteredby the automobile industry in developing and supplying the requiredcontrol technology. The carbon monoxide emission limit for light-duty vehicles at low altitude has been 3.4 g/mi (2.1 g/km) since 1983;since 1984, this limit has applied to light-duty vehicles at all altitudes.

Carbon monoxide emissions from 1990 to 1994 in Germany, asestimated by the Federal Environmental Agency (1997), aresummarized in Table 4 and show a clear reduction with time. Theestimated carbon monoxide emissions in Europe during 1990 aresummarized in Table 5, including the sum of 28 European countriesand individual data from 10 countries (European EnvironmentalAgency, 1995). The Task Group noted that the original reference doesnot specify any details concerning the zero values listed in Table 5.

Table 4. Carbon monoxide national emission estimates: Germanya

Source category 1990 1991 1992 1993 1994

kt % kt % kt % kt % kt %

Industrialprocesses

684 6.4 645 7.1 599 7.6 597 8.1 595 8.8

Production anddistribution of fuel

27 0.3 23 0.3 16 0.2 14 0.2 13 0.2

Street traffic 6 487 60.4 5 593 61.8 4 962 62.6 4 457 60.4 3 953 58.7

Othertransportation

252 2.3 208 2.3 184 2.3 183 2.5 183 2.7

Residential 2 085 19.4 1 512 16.7 1 165 14.7 1 169 15.8 1030 15.3

Small consumers 207 1.9 169 1.9 141 1.8 148 2 143 2.1

Industrialcombustion

871 8.1 775 8.6 745 9.4 705 9.6 716 10.6

Power plants andheating plants

130 1.2 121 1.3 114 1.4 106 1.4 104 1.5

Total 10 743 9 046 7 926 7 379 6 737

a From Federal Environmental Agency (1997).

Table 5. Carbon monoxide emission estimates in Europe in 1990a

Carbon monoxide emission estimates (kt)b

Europe(28 countries)

Aus-tria

France Ger-many

Italy Neth-er-

lands

Po-land

Roma-nia

Spain Swe-den

UnitedKing-dom

kt % oftotal

Public powercogenerationand districtheating

807 1 6 21 466 23 5 68 13 16 6 50

Commercial,institutional andresidentialcombustion

9 947 14 776 1 892 2 053 260 101 1 343 373 890 72 294

Industrialcombustion

8 200 12 27 599 1 174 620 12 3 389 782 406 24 71

Productionprocesses

3 188 6 241 668 664 380 254 122 129 248 6 0

Extraction anddistribution offossil fuels

63 0 0 0 23 0 2 0 0 0 0 2

Solvent use 1 0 0 0 0 0 1 0 0 0 NE 0

Road transport 38 919 56 582 6 812 5 892 5 534 675 2 133 531 2 610 1 118 6 023

Table 5 (contd).

Carbon monoxide emission estimates (kt)b

Europe(28 countries)

Aus-tria

France Ger-many

Italy Neth-er-

lands

Po-land

Roma-nia

Spain Swe-den

UnitedKing-dom

kt % oftotal

Other mobilesourcesand machinery

2 223 3 NE 512 260 719 21 90 27 111 107 42

Wastetreatment anddisposal

4 427 6 0 232 0 1 705 2 225 1 332 527 14 220

Agriculture 579 1 60 NE 0 27 8 0 0 143 0 0

Nature 1 358 2 NE 194 0 1 079 26 18 0 26 2 0

Total 69 712 100 1 692 10 930 10 532 10 347 1 107 7 388 3 187 4 977 1 349 6 702

a From European Environmental Agency (1995).b NE = no estimate.


49

3.2.1 General combustion processes

Incomplete combustion of carbon-containing compounds createsvarying amounts of carbon monoxide. The chemical and physicalprocesses that occur during combustion are complex because theydepend not only on the type of carbon compound reacting withoxygen, but also on the conditions existing in the combustion chamber(Pauling, 1960; Mellor, 1972). Despite the complexity of the combus-tion process, certain general principles regarding the formation ofcarbon monoxide from the combustion of hydrocarbon fuels areaccepted widely.

Gaseous or liquid hydrocarbon fuel reacts with oxygen in a chainof reactions that result in the formation of carbon monoxide. Carbonmonoxide then reacts with hydroxyl radicals to form carbon dioxide.The second reaction is approximately 10 times slower than the first.In coal combustion, too, the reaction of carbon and oxygen to formcarbon monoxide is one of the primary reactions, and a large fractionof carbon atoms go through the carbon monoxide form. Again, theconversion of carbon monoxide to carbon dioxide is much slower.

Four basic variables control the concentration of carbon monox-ide produced in the combustion of all hydrocarbon fuels: (1) oxygenconcentration, (2) flame temperature, (3) gas residence time at hightemperatures and (4) combustion chamber turbulence. Oxygen con-centration affects the formation of both carbon monoxide and carbondioxide, because oxygen is required in the initial reactions with thefuel molecule and in the formation of the hydroxyl radical. As theavailability of oxygen increases, more complete conversion of carbonmonoxide to carbon dioxide results. Flame and gas temperaturesaffect both the formation of carbon monoxide and the conversion ofcarbon monoxide to carbon dioxide, because both reaction ratesincrease exponentially with increasing temperature. Also, thehydroxyl radical concentration in the combustion chamber is verytemperature dependent. The conversion of carbon monoxide to carbondioxide is also enhanced by longer residence time, because this is arelatively slow reaction in comparison with carbon monoxideformation. Increased gas turbulence in the combustion zones increasesthe actual reaction rates by increasing the mixing of the reactants andassisting the relatively slower gaseous diffusion process, therebyresulting in more complete combustion.


50

3.2.2 Combustion engines

3.2.2.1 Mobile combustion engines

Most mobile sources of carbon monoxide are internal combustionengines of two types: (1) gasoline-fuelled, spark ignition,reciprocating engines (carburetted or fuel-injected) and (2) diesel-fuelled reciprocating engines. The carbon monoxide emitted from anygiven engine is the product of the following factors: (1) theconcentration of carbon monoxide in the exhaust gases, (2) the flowrate of exhaust gases and (3) the duration of operation.

1) Internal combustion engines (gasoline-fuelled, spark ignition engines)

Exhaust concentrations of carbon dioxide increase with lower(richer) air-to-fuel (A/F) ratios and decrease with higher (leaner)A/F ratios, but they remain relatively constant with ratios above thestoichiometric ratio of about 15:1 (Hagen & Holiday, 1964). Thebehaviour of gasoline automobile engines before and after theinstallation of pollutant control devices differs considerably. Depend-ing on the mode of driving, the average uncontrolled engine operatesat A/F ratios ranging from about 11:1 to a point slightly above thestoichiometric ratio. During the idling mode, at low speeds with lightload (such as low-speed cruise), during the full open throttle modeuntil speed picks up and during deceleration, the A/F ratio is low inuncontrolled cars, and carbon monoxide emissions are high. Athigher-speed cruise and during moderate acceleration, the reverse istrue. Cars with exhaust controls generally remain much closer tostoichiometric A/F ratios in all modes, and thus the carbon monoxideemissions are kept lower. The exhaust flow rate increases withincreasing engine power output.

The decrease in available oxygen with increasing altitude has theeffect of enriching the A/F mixture and increasing carbon monoxideemissions from carburetted engines. Fuel-injected gasoline engines,which predominate in the vehicle fleet today, have more closelycontrolled A/F ratios and are designed and certified to comply withapplicable emission standards regardless of elevation (US EPA,1983).

Correlations between total emissions of carbon monoxide ingrams per vehicle-mile (vehicle-kilometre) and average route speedshow a decrease in emissions with increasing average speed


51

(Simonaitis & Heicklen, 1972; Stuhl & Niki, 1972; US EPA, 1985).During low-speed conditions (below 32 km/h or 20 mi/h average routespeed), the greater emissions per unit of distance travelled areattributable to (1) an increased frequency of acceleration, decelerationand idling encountered in heavy traffic and (2) the consequentincrease in the operating time per mile (kilometre) driven.

The carbon monoxide and the unburned hydrocarbon exhaustemissions from an uncontrolled engine result from incompletecombustion of the fuel–air mixture. Emission control on new vehiclesis being achieved by engine modifications, improvements in enginedesign and changes in engine operating conditions. Substantialreductions in carbon monoxide and other pollutant emissions resultfrom consideration of design and operating factors such as leaner,uniform mixing of fuel and air during carburetion, controlled heatingof intake air, increased idle speed, retarded spark timing, improvedcylinder head design, exhaust thermal reactors, oxidizing andreducing catalysts, secondary air systems, exhaust recycle systems,electronic fuel injection, A/F ratio feedback controls and modifiedignition systems (NAS, 1973).

2) Internal combustion engines (diesel engines)

Diesel engines are in use throughout the world in heavy-dutyvehicles, such as trucks and buses, and they are also extensively usedin Western Europe in light-duty vans, taxis and some cars. Dieselengines allow more complete combustion and use less volatile fuelsthan do spark ignition engines. The operating principles are signi-ficantly different from those of the gasoline engine. In diesel combus-tion, carbon monoxide concentrations in the exhaust are relatively lowbecause high temperature and large excesses of oxygen are involvedin normal operation.

3.2.2.2 Stationary combustion sources (steam boilers)

This section refers to fuel-burning installations such as coal-,gas- or oil-fired heating or power generating plants (externalcombustion boilers).

In these combustion systems, the formation of carbon monoxideis lowest at a ratio near or slightly above the stoichiometric A/F ratio.At lower than stoichiometric A/F ratios, high carbon monoxide


52

concentrations reflect the relatively low oxygen concentration and thepossibility of poor reactant mixing from low turbulence. These twofactors can increase emissions even though flame temperatures andresidence time are high. At higher than stoichiometric A/F ratios,increased carbon monoxide emissions result from decreased flametemperatures and shorter residence time. These two factors remainpredominant even when oxygen concentrations and turbulenceincrease. Minimal carbon monoxide emissions and maximum thermalefficiency therefore require combustor designs that provide highturbulence, sufficient residence time, high temperatures and near-stoichiometric A/F ratios. Combustor design dictates the actualapproach to that minimum.

3.2.3 Other sources

There are numerous industrial activities that result in theemission of carbon monoxide at one or more stages of the process(Walsh & Nussbaum, 1978; US EPA, 1979a, 1985). Manufacturingpig iron can produce as much as 700–1050 kg carbon monoxide/tonneof pig iron. Other methods of producing iron and steel can producecarbon monoxide at a rate of 9–118.5 kg/tonne. However, most of thecarbon monoxide generated is normally recovered and used as fuel.Conditions such as “slips,” abrupt collapses of cavities in the coke–oremixture, can cause instantaneous emissions of carbon monoxide thattemporarily exceed the capacity of the control equipment. Grey ironfoundries can produce 72.5 kg carbon monoxide/tonne of product, butan efficient afterburner can reduce the carbon monoxide emissions to4.5 kg/tonne. Nevertheless, industrial carbon monoxide emissionsmay constitute an important part of total emissions in industrial cities— for example, in the Ruhr area in Germany.

Charcoal production results in average carbon monoxideemissions of 172 kg/tonne. Emissions from batch kilns are difficult tocontrol, although some may have afterburners. Afterburners can moreeasily reduce, by an estimated 80% or more, the relatively constantcarbon monoxide emissions from continuous charcoal production.Emissions from carbon black manufacture can range from 5 to3200 kg carbon monoxide/tonne depending on the efficiency andquality of the emission control systems.

Some chemical processes, such as phthalic anhydride production,give off as little as 6 kg carbon monoxide/tonne with proper controls


53

or as much as 200 kg carbon monoxide/tonne if no controls areinstalled. There are numerous other chemical processes that producerelatively low carbon monoxide emissions per tonne of product:sulfate pulping for paper produces 1–30 kg carbon monoxide/tonne,lime manufacturing normally produces 1–4 kg carbon monoxide/tonne, and carbon monoxide emissions from adipic acid productionare zero or slight with proper controls. Other industrial chemicalprocesses that cause carbon monoxide emissions are the manufactureof terephthalic acid and the synthesis of methanol and higheralcohols. As a rule, most industries find it economically desirable toinstall suitable controls to reduce carbon monoxide emissions.

Even though some of these carbon monoxide emission rates seemexcessively high, they are, in fact, only a small part of the totalpollutant load. Mention of these industries is made to emphasize theconcern for localized pollution problems when accidents occur orproper controls are not used.

In some neighbourhoods, wintertime carbon monoxide emissionsinclude a significant component from residential fireplaces and woodstoves. Emissions of carbon monoxide can range from 18 to 140 g/kg,depending on design, fuel type and skill of operation.

Although the estimated carbon monoxide emissions resultingfrom forest wildfires in the USA have fluctuated between about 4 and9 million tonnes per year since 1970 and were 6.2 million tonnes in1989, the estimated total carbon monoxide emissions from industrialprocesses in the USA declined from 8.9 million tonnes in 1970 to 4.6million tonnes in 1989 (US EPA, 1991c).

3.3 Indoor carbon monoxide

3.3.1 Introduction

Carbon monoxide is introduced to indoor environments throughemissions from a variety of combustion sources and in the infiltrationor ventilation air from outdoors. The resulting indoor concentration,both average and peak, is dependent on a complex interaction ofseveral interrelated factors affecting the introduction, dispersion andremoval of carbon monoxide. These factors include, for example, suchvariables as (1) the type, nature (factors affecting the generation rate


54

of carbon monoxide) and number of sources, (2) source usecharacteristics, (3) building characteristics, (4) infiltration orventilation rates, (5) air mixing between and within compartments inan indoor space, (6) removal rates and potential remission orgeneration by indoor surfaces and chemical transformations,(7) existence and effectiveness of air contaminant removal systemsand (8) outdoor concentrations.

Source emissions from indoor combustion are usually charac-terized in terms of emission rates, defined as the mass of pollutantemitted per unit of fuel input (micrograms per kilojoule). Theyprovide source strength data as input for indoor modelling, promotean understanding of the fundamental processes influencing emissions,guide field study designs assessing indoor concentrations, identify andrank important sources and aid in developing effective mitigationmeasures. Unfortunately, source emissions can vary widely. Althoughit would be most useful to assess the impact of each of the sources onindoor air concentrations of carbon monoxide by using models, thehigh variability in the source emissions and in other factors affectingthe indoor levels does not make such an effort very useful. Such anestimate will result in predicted indoor concentrations ranging overseveral orders of magnitude, making them of no practical use, andmay be misleading.

3.3.2 Emissions from indoor sources

Carbon monoxide emitted directly into the indoor environmentis one of several air contaminants resulting from combustion sources.Such emissions into occupied spaces can be unintentional or the resultof accepted use of unvented or partially vented combustion sources.Faulty or leaky flue pipes, backdrafting and spillage from combustionappliances that draw their air from indoors (e.g., Moffatt, 1986),improper use of combustion sources (e.g., use of a poorly maintainedkerosene heater) and air intake into a building from attached parkinggarages are all examples of unintentional or accidental indoor sourcesof carbon monoxide. In the USA, the National Center for HealthStatistics (1986) estimates that between 700 and 1000 deaths per yearare due to accidental carbon monoxide poisoning. Mortality statisticsare similar for other developed countries as well. The number ofindividuals experiencing severe adverse health effects at sublethalcarbon monoxide concentrations from accidental indoor sources is nodoubt many times the number of estimated deaths. Although the


55

unintentional or accidental indoor sources of carbon monoxiderepresent a serious health hazard, little is known about the extent ofthe problem throughout the world. Such sources cannot becharacterized for carbon monoxide emissions in any standard way thatwould make the results extendable to the general population.

The major indoor sources of carbon monoxide emissions thatresult from the accepted use of unvented or partially vented com-bustion sources include gas cooking ranges and ovens, gas appliances,unvented gas space heaters, unvented kerosene space heaters, coal- orwood-burning stoves and cigarette combustion.

3.3.2.1 Gas cooking ranges, gas ovens and gas appliances

Estimates indicate that gas (natural gas and liquid propane) isused for cooking, heating water and drying clothes in approximately45.1% of all homes in the USA (US Bureau of the Census, 1982) andin nearly 100% of the homes in some other countries (e.g., theNetherlands). Unvented, partially vented and improperly vented gasappliances, particularly the gas cooking range and oven, represent animportant source category of carbon monoxide emissions into theindoor residential environment. Emissions of carbon monoxide fromthese gas appliances are a function of a number of variables relatingto the source type (range top or oven, water heater, dryer, number ofpilot lights, burner design, etc.), source condition (age, maintenance,combustion efficiency, etc.), source use (number of burners used,frequency of use, fuel consumption rate, length of use, improper use,etc.) and venting of emissions (existence and use of outside vents overranges, efficiency of vents, venting of gas dryers, etc.).

The source emission studies typically have been conducted in thelaboratory setting and have involved relatively few gas ranges and gasappliances. The reported studies indicate that carbon monoxideemissions are highly variable among burners on a single gas cookingrange and between gas cooking ranges and ovens, varying by as muchas an order of magnitude. Operating a gas cooking range or ovenunder improperly adjusted flame conditions (yellow-tipped flame) canresult in greater than a fivefold increase in emissions compared withproperly operating flame conditions (blue flame). Use of a rich or leanfuel appeared to have little effect on carbon monoxide emissions. Ingeneral, carbon monoxide emissions were roughly, on average,comparable for top burners, ovens, pilot lights and unvented gas


56

dryers when corrected for fuel consumption rate. The emission ratesgathered by either the direct or mass balance method werecomparable. Only one study attempted to evaluate gas stove emissionsin the field for a small number (10) of residences. This study foundcarbon monoxide emissions to be as much as a factor of 4 higher thanin chamber studies. Given the prevalence of the source, limited fieldmeasurements and poor agreement between existing laboratory- andfield-derived carbon monoxide emission data, there is a need toestablish a better carbon monoxide emission database for gas cookingranges in residential settings.

3.3.2.2 Unvented space heaters

Unvented kerosene and gas space heaters are used in the colderclimates to supplement central heating systems or in more moderateclimates as the primary source of heat. During the heating season,space heaters generally will be used for a number of hours during theday, resulting in emissions over relatively long periods of time.

Over the last several years, there has been a dramatic increase inthe use of unvented or poorly vented kerosene space heaters inresidential and commercial establishments, primarily as a supplemen-tal heat source. For example, in the USA, an estimated 16.1 millionsuch heaters had been sold through 1986 (S.E. Womble, personalcommunication, US Consumer Product Safety Commission, 1988).An additional 3 million residences use unvented gas space heaters(fuelled by natural gas or propane). The potentially large number ofunvented space heaters used throughout the world, particularly duringperiods when energy costs rise quickly, makes them an importantsource of carbon monoxide indoors.

Carbon monoxide emissions from unvented kerosene and gasspace heaters can vary considerably and are a function of heaterdesign (convective, radiant, combination, etc.), condition of heaterand manner of operation (e.g., flame setting).

Carbon monoxide emissions from unvented gas space heaterswere found to be variable from heater to heater, but were roughlycomparable to those for gas cooking ranges. Infrared gas space heatersproduced higher emissions than the convective or catalytic heaters.Emissions of carbon monoxide for these heaters were higher formaltuned heaters and for the mass balance versus direct method of


57

testing. No differences for rich or lean fuel were found, but use ofnatural gas resulted in higher emissions than did use of propane.Lower fuel consumption settings resulted in lower carbon monoxideemissions. Emissions were observed to vary in time during a heaterrun and increase when room or chamber oxygen levels decreased.

Among the three principal unvented kerosene space heaterdesigns (radiant, convective and two-stage burners), radiant heatersproduced the highest carbon monoxide emissions and convectiveheaters produced the lowest emissions. Wick setting (low, normal orhigh) had a major impact on emissions, with the low-wick settingresulting in the highest carbon monoxide emissions. Data fromdifferent laboratories are in good agreement for this source.

3.3.2.3 Coal or wood stoves

Use of coal- or wood-burning stoves has been a popular costsavings alternative to conventional cooking and heating systems usinggas or oil. Carbon monoxide and other combustion by-products enterthe indoor environment during fire start-up, during fire-tendingfunctions or through leaks in the stove or venting system. Hence, it isdifficult to evaluate indoor carbon monoxide emission rates for thesesources. Traynor et al. (1987) evaluated indoor carbon monoxidelevels from four wood-burning stoves (three airtight stoves and onenon-airtight stove) in a residence. The non-airtight stove emittedsubstantial amounts of carbon monoxide to the residence, particularlywhen operated with a large fire. The airtight stoves contributedconsiderably less. The average carbon monoxide source strengthsduring stove operation ranged from 10 to 140 cm3/h for the airtightstoves and from 220 to 1800 cm3/h for the non-airtight stoves.

3.3.2.4 Tobacco combustion

The combustion of tobacco represents an important source ofindoor air contaminants. Carbon monoxide is emitted indoors fromtobacco combustion through the exhaled mainstream smoke and fromthe smouldering end of the cigarette (sidestream smoke). Carbonmonoxide emission rates in mainstream and sidestream smoke havebeen evaluated extensively in small chambers (less than a litre involume) using a standardized smoking machine protocol. The resultsof these studies have been summarized and evaluated in severalreports (e.g., NRC, 1986a; Surgeon General of the United States,


58

1986). These results indicate considerable variability in total(mainstream plus sidestream smoke) carbon monoxide emissions,with a typical range of 40–67 mg per cigarette. A small chamberstudy of 15 brands of Canadian cigarettes (Rickert et al., 1984) foundthe average carbon monoxide emission rate (mainstream plussidestream smoke) to be 65 mg per cigarette. A more limited numberof studies have been done using large chambers with the occupantssmoking or using smoking machines. Girman et al. (1982) reporteda carbon monoxide emission rate of 94.6 mg per cigarette for a largechamber study in which one cigarette brand was evaluated. A carbonmonoxide emission factor of 88.3 mg per cigarette was reported byMoschandreas et al. (1985) for a large chamber study of one referencecigarette.

On average, a smoker smokes approximately two cigarettes perhour, with an average smoking time of approximately 10 min percigarette. Using the above range of reported carbon monoxideemission rates for environmental tobacco smoke, this would roughlyresult in the emission of 80–190 mg of carbon monoxide per smokerper hour into indoor spaces where smoking occurs. This valuecompares with an approximate average carbon monoxide emissionrate of 260–545 mg/h for one range-top burner (without pilot light)operating with a blue flame. Two smokers in a house would producehourly carbon monoxide emissions comparable to the hourlyproduction rate of a single gas burner. Tobacco combustion thereforerepresents an important indoor source of carbon monoxide,particularly in locations where many people are smoking.

59

4. ENVIRONMENTAL DISTRIBUTION ANDTRANSFORMATION

4.1 Introduction

Carbon monoxide was first discovered to be a minor constituentof the Earth’s atmosphere in 1948 by Migeotte (1949). While takingmeasurements of the solar spectrum, he observed a strong absorptionband in the infrared region at 4.6 :m, which he attributed to carbonmonoxide (Lagemann et al., 1947). On the twin bases of the beliefthat the solar contribution to that band was negligible and hisobservation of a strong day-to-day variability in absorption, Migeotte(1949) concluded that an appreciable amount of carbon monoxide waspresent in the terrestrial atmosphere of Columbus, Ohio, USA. In the1950s, many more observations of carbon monoxide were made, withmeasured concentrations ranging from 0.09 to 110 mg/m3 (0.08 to100 ppm) (Migeotte & Neven, 1952; Benesch et al., 1953; Locke &Herzberg, 1953; Faith et al., 1959; Robbins et al., 1968; Sie et al.,1976). On the basis of these and other measurements available in1963, Junge (1963) stated that carbon monoxide appeared to be themost abundant trace gas, other than carbon dioxide, in theatmosphere. The studies of Sie et al. (1976) indicated higher mixingratios near the ground than in the upper atmosphere, implying asource in the biosphere, but Junge (1963) emphasized that knowledgeof the sources and sinks of atmospheric carbon monoxide wasextremely poor. It was not until the late 1960s that concerted effortswere made to determine the various production and destructionmechanisms for carbon monoxide in the atmosphere.

Far from human habitation in remote areas of the southern hemi-sphere, natural background carbon monoxide concentrations averagearound 0.05 mg/m3 (0.04 ppm), primarily as a result of natural pro-cesses such as forest fires and the oxidation of methane. In the nor-thern hemisphere, background concentrations are 2–3 times higherbecause of more extensive human activities. Much higher concentra-tions occur in cities, arising from technological sources such as auto-mobiles and the production of heat and power. Carbon monoxideemissions are increased when fuel is burned in an incomplete orinefficient way.


60

The physical and chemical properties of carbon monoxidesuggest that its atmospheric removal occurs primarily by reaction withhydroxyl radicals. Almost all the carbon monoxide emitted into theatmosphere each year is removed by reactions with hydroxyl radicals(85%), by soils (10%) and by diffusion into the stratosphere. There isa small imbalance between annual emissions and removal, causing anincrease of about 1% per year. It is very likely that the imbalance isdue to increasing emissions from anthropogenic activities. The aver-age concentration of carbon monoxide is about 100 :g/m3 (90 ppbv),which amounts to about 400 million tonnes in the atmosphere, andthe average lifetime is about 2 months. This view of the global cycleof carbon monoxide is consistent with the present estimates of averagehydroxyl radical concentrations and the budgets of other trace gases,including methane and methyl chloroform.

4.2 Global sources, sinks and lifetime

The largest sources of carbon monoxide in the global atmosphereare combustion processes and the oxidation of hydrocarbons (seechapter 3). The mass balance of a trace gas in the atmosphere can bedescribed as a balance between the rate of change of the global burdenadded to the annual rate of loss on the one side and global emissionson the other side (dC/dt + loss rate = source emissions, where C =concentration). In steady state, the atmospheric lifetime (J) is the ratioof the global burden to the loss rate. The global burden is the totalnumber of molecules of a trace gas in the atmosphere or its total mass.The concentration of a trace gas can vary (dC/dt is not 0) when eitherthe loss rate or the emissions vary cyclically in time, representingseasonal variations, or vary over a long time, often representing trendsin human industrial activities or population. For carbon monoxide,both types of trends exist. There are large seasonal cycles drivenmostly by seasonal variations in the loss rate but also affected byseasonal variations in emissions, and there are also indications oflong-term trends probably caused by increasing anthropogenicemissions.

4.2.1 Sinks

It is believed that reaction with hydroxyl radicals is the majorsink for removing carbon monoxide from the atmosphere. The cycleof the hydroxyl radical itself cannot be uncoupled from the cycles ofcarbon monoxide, methane, water and ozone. In the troposphere,

Environmental Distribution and Transformation

61

hydroxyl radicals (OHA) are produced by the photolysis of ozone (hL +O3 6 O(1D) + O2) followed by the reaction of the excited oxygenatoms with water vapour to produce two hydroxyl radicals (O(1D) +H2O 6 OHA + OHA). The production of hydroxyl radicals is balancedby their removal principally by reactions with carbon monoxide andmethane. On a global scale, carbon monoxide may remove morehydroxyl radicals than methane; however, methane is more importantin the southern hemisphere, where there is much less carbon monox-ide, than in the northern hemisphere, but the amount of methane isonly slightly less in the northern hemisphere.

The amount of carbon monoxide that is removed by reactionswith hydroxyl radicals can be estimated by calculating the loss asloss = Keff [OHA]ave [CO]ave, where Keff is the effective reaction rateconstant, [OHA]ave is the average hydroxyl radical concentration and[CO]ave is the average concentration of carbon monoxide. Thereaction rate constant of CO + OHA is K = (1.5 × 10–13) (1 + 0.6 Patm)cm3/molecule per second (DeMore et al., 1987), where Patm is theatmospheric pressure. The constant Keff describes the effective reactionrate, taking into account the decreasing atmospheric pressure anddecreasing carbon monoxide concentrations with height. EstimatingKeff to be 2 × 10–13 cm3/molecule per second and taking [OHA]ave to be8 × 105 molecules/cm3 and [CO]ave to be 90 ppbv (equivalent to about100 :g/m3), the annual loss of carbon monoxide from reactions withhydroxyl radicals is about 2200 million tonnes per year. The valuesadopted for [OHA]ave and [CO]ave are discussed in more detail later inthis chapter.

Uptake of carbon monoxide by soils has been documented andmay amount to about 250 million tonnes per year, or about 10% of thetotal emitted into the atmosphere (Inman et al., 1971; Ingersoll et al.,1974; Seiler & Schmidt, 1974; Bartholomew & Alexander, 1981),although arid soils may release carbon monoxide into the atmosphere(Conrad & Seiler, 1982). Another 100 million tonnes (5%) or so areprobably removed annually in the stratosphere (Seiler, 1974).

4.2.2 Atmospheric lifetime

Based on the global sources and sinks described above, theaverage atmospheric lifetime of carbon monoxide can be calculated tobe about 2 months, with a range between 1 and 4 months, whichreflects the uncertainty in the annual emissions of carbon monoxide


62

(J = C/S, where C is the tropospheric mixing ratio and S is the totalannual emissions). The lifetime, however, can vary enormously withlatitude and season compared with its global average value. Duringwinters at high and middle latitudes, carbon monoxide has a lifetimeof more than a year, but during summers at middle latitudes, thelifetime may be closer to the average global lifetime of about2 months. Moreover, in the tropics, the average lifetime of carbonmonoxide is probably about 1 month. These calculated variationsreflect the seasonal cycles of hydroxyl radicals at various latitudes.

4.2.3 Latitudinal distribution of sources

When the sources, sinks, transport and observed concentrationsof carbon monoxide are combined into a mass balance model, it ispossible to calculate any one of these four components if the others areknown. In the case of carbon monoxide, the sources can be estimatedassuming that the sinks (hydroxyl radical reaction and soils),transport and concentrations are known. The latitudinal distributionof sources can be described in a one-dimensional model (Khalil &Rasmussen, 1990b). This model is similar to that described byCzeplak & Junge (1974) and Fink & Klais (1978). A time-averagedversion was applied to the carbon monoxide budget by Hameed &Stewart (1979), and a somewhat modified and time-dependentversion, mentioned above, was applied by Khalil & Rasmussen(1990b) to derive the latitudinal distribution of carbon monoxideshown in Fig. 2. Calculations by Khalil & Rasmussen (1990b) alsosuggest that emissions are higher in spring and summer than in theother seasons, particularly in the middle northern latitudes. This isexpected for three reasons: (1) oxidation of methane and otherhydrocarbons is faster during the summer because of the seasonalvariation of hydroxyl radicals; (2) other direct emissions are alsogreater during spring and summer; and (3) at middle and higherlatitudes, methane and non-methane hydrocarbons build up during thewinter, and this reservoir is oxidized when hydroxyl radicalconcentrations rise during the spring.

From Fig. 2, the emissions from the northern and southern tropi-cal latitudes sum up to 480 million tonnes per year and 330 milliontonnes per year, respectively; the emissions from the northern andsouthern middle latitudes are 960 million tonnes per year and210 million tonnes per year, respectively; some 50 million tonnes areemitted each year from the Arctic; and some 10 million tonnes peryear come from the Antarctic. The largest fluxes of carbon monoxide


63

Fig. 2. The estimated emissions of carbon monoxide as a function of latitude. Theemissions are in million tonnes/year (Mt/year) in each latitude band 0.02 units in sineof latitude. The dashed lines are estimates of uncertainties as hydroxyl radicalconcentrations and the rate of dispersion are varied simultaneously so that themaximum values of each of these parameters are twice the minimum values (fromKhalil & Rasmussen, 1990b).

are from the industrial band of latitudes between 30 and 50 °N. Fromthis region, some 620 million tonnes per year are emitted,representing about 30% of the total emissions of 2050 million tonnesper year. The model does not distinguish between anthropogenic andnatural sources, nor does it distinguish between direct emissions andphotochemical production of carbon monoxide from the oxidation ofhydrocarbons. A large part of the estimated fluxes from the mid-northern latitudes and from tropical regions is likely to be of anthro-pogenic origin. The latitudinal distribution in Fig. 2 is compatiblewith the estimate (from Table 3 in chapter 3) that about 60% of thetotal carbon monoxide emissions are from anthropogenic activities.

4.2.4 Uncertainties and consistencies

The first consistency one notes is that the total emissions ofcarbon monoxide estimated from the various sources are balanced bythe estimated removal of carbon monoxide. The approximate balancebetween sources and sinks is expected because the trends are showingan increase of only about 4–8 million tonnes per year compared with


64

the total global emission rate of more than 2000 million tonnes peryear.

On the other hand, there are large uncertainties in the sourcesand sinks that in the future may upset the apparently cohesive presentbudget of carbon monoxide. Although the patterns of the globaldistribution are becoming established, there are still uncertaintiesabout the absolute concentrations. Estimates of emissions fromindividual sources are very uncertain (see Table 3 in chapter 3). Inmost cases, the stated uncertainty is a qualitative expression of thelikely range of emissions, and it cannot be interpreted statistically.Therefore, the resulting uncertainty in the total emissions, obtainedby adding up the uncertainties in individual sources, appears to belarge.

There are two difficulties encountered in improving the estimatesof carbon monoxide emissions from individual sources. First,although many critical experiments to determine the production andemissions of carbon monoxide from individual sources are yet to bedone, there is a limit to the accuracy with which laboratory data canbe extrapolated to the global scale. Second, the cycle of carbonmonoxide may be so intimately tied up with the cycles ofhydrocarbons that accurate global estimates of carbon monoxideemissions may not be possible until the hydrocarbon cycles are betterunderstood.

Whereas the global distribution and seasonal variations in theglobal distribution of hydroxyl radicals can be calculated, there are nodirect measurements of hydroxyl radicals that can be used to estimatethe removal of carbon monoxide. The effective average concentrationof hydroxyl radicals that acts on trace gases can be estimatedindirectly from the cycles of other trace gases with known globalemissions. Therefore, the total emissions of carbon monoxide areconstrained by the budgets of other trace gases, even though theestimates of emissions from individual sources may remain uncertain.The most notable constraint may be the budget of methyl chloroform.Methyl chloroform is a degreasing solvent that has been emitted intothe atmosphere in substantial quantities for more than 20 years. It isthought to be removed principally by reacting with hydroxyl radicalsand to a lesser extent by photodissociation in the stratosphere.Because industry records on methyl chloroform production and saleshave been kept for a long time, methyl chloroform can be used toestimate the average amount of hydroxyl radicals needed to explainthe observed concentrations compared with the emissions. The


65

accuracy of the source estimates of methyl chloroform is improved bythe patterns of its uses; most methyl chloroform tends to be releasedshortly after purchase, so large unknown or unquantified reservoirsprobably do not exist. The recent budgets of methyl chloroformsuggest that, on average, there are about 8 × 105 molecules ofhydroxyl radical per cubic centimetre, although significantuncertainties remain (see, for example, Khalil & Rasmussen, 1984c).This is the value used above in estimating the loss of carbonmonoxide from reaction with hydroxyl radicals. The same averagevalue of hydroxyl radicals also explains the methane concentrationscompared with estimated sources, lending more support to theaccuracy of the estimated hydroxyl radical concentrations. Neither ofthese constraints is very stringent; however, if the total globalemissions of carbon monoxide from all sources are much differentfrom the estimated 2600 million tonnes per year, then revisions of thebudgets of both methane and methyl chloroform may be required.

Although there are other sources and sinks of carbon monoxide,these are believed to be of lesser importance on a global scale (Swin-nerton et al., 1971; Chan et al., 1977).

4.3 Global distributions

Atmospheric concentrations, and thus the global distribution, aregenerally the most accurately known components of a global massbalance of a trace gas, because direct atmospheric measurements canbe taken (Wilkniss et al., 1973; Seiler, 1974; Ehhalt & Schmidt,1978; Pratt & Falconer, 1979; Heidt et al., 1980; Dianov-Klokov &Yurganov, 1981; Seiler & Fishman, 1981; Rasmussen & Khalil,1982; Reichle et al., 1982; Hoell et al., 1984; Fraser et al., 1986;Khalil & Rasmussen, 1988, 1990a). Much has been learned about theglobal distribution of carbon monoxide over the last decade. Theexperiments leading to the present understanding range fromsystematic global observations at ground level for the last 8–10 years,reported by Khalil & Rasmussen (1988, 1990a) and Seiler (Seiler &Junge, 1970; Seiler, 1974), to finding the instantaneous globaldistribution of carbon monoxide from remote-sensing instruments onboard the US National Aeronautics and Space Administration’s spaceshuttle, as reported by Reichle et al. (1982, 1990).


66

4.3.1 Seasonal variations

The seasonal variations in carbon monoxide are well established(Dianov-Klokov & Yurganov, 1981; Seiler et al., 1984; Fraser et al.,1986; Khalil & Rasmussen, 1990a). High concentrations are observedduring the winters in each hemisphere, and the lowest concentrationsare seen in late summer. The amplitude of the cycle is largest at highnorthern latitudes and diminishes as one moves towards the equatoruntil it is reversed in the southern hemisphere, reflecting the reversalof the seasons. The seasonal variations are small in the equatorialregion. These patterns are expected from the seasonal variations inhydroxyl radical concentrations and carbon monoxide emissions. Atmid and high latitudes, diminished solar radiation, water vapour andozone during winters cause the concentrations of hydroxyl radicals tobe much lower than during summer. The removal of carbon monoxideis slowed down, and its concentrations build up. In summer, theopposite pattern exists, causing the large seasonal variations in carbonmonoxide.

On the hemispheric scale, the seasonal variation in carbon mon-oxide is approximately proportional to the concentration. Therefore,because there is much more carbon monoxide in the northern hemi-sphere than in the southern hemisphere, the decline of concentrationsin the northern hemisphere during the summer is not balanced by therise of concentrations in the southern hemisphere. This causes aglobal seasonal variation. The total amount of carbon monoxide in theEarth’s atmosphere undergoes a remarkably large seasonal variation;the global burden is highest during northern winters and lowestduring northern summers.

4.3.2 Latitudinal variation

The global seasonal variation in carbon monoxide content in theEarth’s atmosphere also creates a seasonal variation in the latitudinaldistribution (Newell et al., 1974; Seiler, 1974; Reichle et al., 1982,1986; Khalil & Rasmussen, 1988, 1990a). During northern winters,carbon monoxide levels are at their highest in the northern hemi-sphere, whereas concentrations in the southern hemisphere are at aminimum. The interhemispheric gradient, defined as the ratio of theamounts of carbon monoxide in the northern and southern hemi-spheres, is at its maximum of about 3.2 during northern hemispherewinters and falls to about 1.8 during northern hemisphere summers,


67

which is about half the winter value. The average latitudinal gradientis about 2.5, which means that, on average, there is about 2.5 times asmuch carbon monoxide in the northern hemisphere as in the southernhemisphere. Early data on latitudinal variations did not account forthe seasonal variations.

4.3.3 Variations with altitude

In the northern hemisphere troposphere, the concentrations ofcarbon monoxide generally decline with altitude, but in the southernhemisphere, the vertical gradient may be reversed as a result of thetransport of carbon monoxide from the northern hemisphere into thesouthern hemisphere. Above the tropopause, concentrations declinerapidly, so that there is very little carbon monoxide between 20 and40 km; at still higher altitudes, the mixing ratio may again increase(Seiler & Junge, 1969; Seiler & Warneck, 1972; Fabian et al., 1981).

4.3.4 Other variations

The concentration of carbon monoxide is generally higher in theair over populated continental areas than in the air over oceans, eventhough oceans release carbon monoxide into the atmosphere. Otherregions, such as tropical forests, may also be a source of isoprene andother hydrocarbons that may form carbon monoxide in the atmos-phere. Such sources produce shifting patterns of high carbonmonoxide concentrations over regional and perhaps even largerspatial scales. Variations in carbon monoxide concentrations weremeasured during the 1984 flights of the space shuttle, as reported byReichle et al. (1990).

Occasionally, significant diurnal variations in carbon monoxideconcentrations may also occur in some locations. For instance, diurnalvariations have been observed over some parts of the oceans, withhigh concentrations during the day and low concentrations at night.Because similar patterns also exist in the surface seawater, the diurnalvariations in carbon monoxide concentrations in the air can beexplained by emissions from the oceans.

Finally, after the repeating cycles and other trends are subtracted,considerable random fluctuations still remain in time series ofmeasurements. These fluctuations reflect the short lifetime of carbon


68

monoxide and the vicinity of the sources, and they complicate thedetection of long-term trends.

4.4 Global trends

Global concentrations of carbon monoxide were reported to beincreasing during the late 1970s and early 1980s, because some 60%of the global emissions of carbon monoxide come from anthropogenicsources, which had increasing emissions over this period. Directatmospheric observations reported by Khalil & Rasmussen (1984a)showed a detectable increasing trend at Cape Meares in Oregon, USA,between 1979 and 1982, when the rate of increase was about 5% peryear. Subsequent data from the same site showed that the rate was notsustained for long, and a much smaller increasing trend of somewhatless than 2% per year emerged over the longer period of 1970–1987(Khalil & Rasmussen, 1988). Similar data from other sites distributedworldwide also showed a global increase of about 1% per year (Khalil& Rasmussen, 1988). The trends were strongest in the mid-northernlatitudes where most of the sources were located and became smallerand weaker in the southern hemisphere. At the mid-southern latitudesite, the trends persisted but were not statistically significant (Khalil& Rasmussen, 1988). Rinsland & Levine (1985) reported estimates ofcarbon monoxide concentrations from spectroscopic plates fromEurope showing that between 1950 and 1984, carbon monoxideconcentrations increased at about 2% per year. Spectroscopicmeasurements of carbon monoxide taken by Dvoryashina et al. (1982,1984) and Dianov-Klokov and colleagues (Dianov-Klokov et al.,1978; Dianov-Klokov & Yurganov, 1981) in the Soviet Union alsosuggested an increase of about 2% per year between 1974 and 1982(Khalil & Rasmussen, 1984b, 1988).

More recent reports of carbon monoxide measurements in airsamples show that from 1988 to 1993, global carbon monoxideconcentrations started to decline rapidly. Novelli et al. (1994) col-lected air samples from 27 locations between 71 °N and 41 °S aboutonce every 3 weeks from a ship during the period June 1990 to June1993. In the northern latitudes, carbon monoxide concentrationsdecreased at a spatially and temporally average rate of 8.4 ± 1.0:g/m3 (7.3 ± 0.9 ppb) per year (6.1% per year). In the southernlatitudes, carbon monoxide concentrations decreased at a rate of 4.8± 0.6 :g/m3 (4.2 ± 0.5 ppb) per year (7.0% per year). Khalil &Rasmussen (1994) reported a slightly smaller decline in global carbon


69

monoxide concentrations of 2.6 ± 0.8% per year during the periodfrom 1988 to 1992. The rate of decrease reported by Khalil &Rasmussen (1994) was particularly rapid in the southern hemisphere.The authors hypothesize that this decline may reflect a reduction intropical biomass burning.

Since 1993, the downward trend in global carbon monoxideconcentrations has levelled off, and it is not clear if carbon monoxidewill continue to decline or increase. These reported trends in globalcarbon monoxide concentrations are relatively small, and the randomvariability is large. Nevertheless, they are extremely importanttowards an understanding of global atmospheric chemistry andpossible effects on global climate. Such changes in troposphericcarbon monoxide concentrations can cause shifts in hydroxyl radicalconcentrations and affect the oxidizing capacity of the atmosphere,thereby influencing the concentration of other trace gases, includingmethane. This change could also be a significant factor contributingto levels of ozone in the non-urban troposphere.

The likely future global-scale concentrations of carbon monoxideare completely unknown at present. It is possible that in the nextdecade, carbon monoxide concentrations will remain stable or evendecline further. Emissions from automobiles are probably on thedecline worldwide, emissions from biomass burning may bestabilizing or even declining, as speculated above, and thecontribution from methane oxidation may no longer be increasing asrapidly as before. Because the atmospheric lifetime of carbonmonoxide is short compared with those of other contributors to globalchange, the ambient concentrations adjust rapidly to existingemissions of carbon monoxide or its precursors.

70

5. ENVIRONMENTAL LEVELS AND PERSONALEXPOSURES

5.1 Introduction

Air quality guidelines for carbon monoxide are designed toprotect against actual and potential human exposures in ambient airthat would cause adverse health effects. The World Health Organi-zation’s guidelines for carbon monoxide exposure (WHO, 1987) areexpressed at four averaging times, as follows:

100 mg/m3 for 15 min60 mg/m3 for 30 min30 mg/m3 for 1 h10 mg/m3 for 8 h

The guideline values and periods of time-weighted average exposureshave been determined so that the carboxyhaemoglobin level of 2.5%is not exceeded, even when a normal subject engages in relativelyheavy work.

Cigarette consumption represents a special case of carbonmonoxide exposure; for the smoker, it almost always dominates overpersonal exposure from other sources. Studies by Radford & Drizd(1982) show that carboxyhaemoglobin levels of cigarette smokersaverage 4%, whereas those of non-smokers average 1%. Therefore,this summary focuses on environmental exposure of non-smokers tocarbon monoxide.

People encounter carbon monoxide in a variety of environments— while travelling in motor vehicles, working at their jobs, visitingurban locations associated with combustion sources, or cooking andheating with domestic gas, charcoal or wood fires — as well as intobacco smoke. Studies of human exposure have shown that amongthese settings, the motor vehicle is the most important for regularlyencountered elevations of carbon monoxide. Studies conducted by theUS EPA in Denver, Colorado, and Washington, DC, for example,have demonstrated that the motor vehicle interior has the highestaverage carbon monoxide concentrations (averaging 8–11 mg/m3

[7–10 ppm]) of all microenvironments (Johnson, 1984).

Environmental Levels and Personal Exposures

71

Another important setting for carbon monoxide exposure is theworkplace. In general, carbon monoxide exposures at work exceedexposures during non-work periods, apart from commuting to andfrom work. Average concentrations may be elevated during this periodbecause workplaces are often located in congested areas that havehigher background carbon monoxide concentrations than do manyresidential neighbourhoods. Occupational and non-occupational expo-sures may overlay one another and result in a higher concentration ofcarbon monoxide in the blood. Certain occupations also increase therisk of high carbon monoxide exposure. These include those occupa-tions involved directly with vehicle driving, maintenance and parking,such as auto mechanics; parking garage and gas station attendants;bus, truck or taxi drivers; traffic police; and warehouse workers. Someindustrial processes produce carbon monoxide directly or as a by-product, including steel production, nickel refining, coke ovens,carbon black production and petroleum refining. Firefighters, cooksand construction workers may also be exposed to higher carbon mon-oxide levels at work. Occupational exposures in industries or settingswith carbon monoxide production also represent some of the highestindividual exposures observed in field monitoring studies.

The highest indoor non-occupational carbon monoxide exposuresare associated with combustion sources and include enclosed parkinggarages, service stations and restaurants. The lowest indoor carbonmonoxide concentrations are found in homes, churches and healthcare facilities. The US EPA’s Denver Personal Monitoring Studyshowed that passive cigarette smoke is associated with increasing anon-smoker’s exposure by an average of about 1.7 mg/m3 (1.5 ppm)and that use of a gas range is associated with an increase of about2.9 mg/m3 (2.5 ppm) at home. Other sources that may contribute tohigher carbon monoxide levels in the home include combustion spaceheaters and wood- and coal-burning stoves.

5.2 Population exposure to carbon monoxide

5.2.1 Ambient air monitoring

Many early attempts to estimate exposure of human populationsused data on ambient air quality from fixed monitoring stations. Anexample of such an analysis can be found in the 1980 annual reportof the President’s Council on Environmental Quality (1980). In this


72

analysis, a county’s exposure to an air pollutant was estimated as theproduct of the number of days on which violations of the primarystandard were observed at county monitoring sites multiplied by thecounty’s population. Exposure was expressed in units of person-days.National exposure to an air pollutant was estimated by the sum of allcounty exposures.

The methodology employed by the Council on EnvironmentalQuality provides a relatively crude estimate of exposure and is limitedby four assumptions:

(1) The exposed populations do not travel outside areas repre-sented by fixed-site monitors.

(2) The air pollutant concentrations measured with the networkof fixed-site monitors are representative of the concentra-tions breathed by the population throughout the area.

(3) The air quality in any one area is only as good as that at thelocation that has the worst air quality.

(4) There are no violations in areas of the county that are notmonitored.

Many studies cast doubt on the validity of these assumptions forcarbon monoxide. These studies are reviewed in Ott (1982) and inSpengler & Soczek (1984). Doubts over the ability of fixed-sitemonitors alone to accurately depict air pollutant exposures are basedon two major findings on fixed-site monitor representativeness:

(1) Indoor and in-transit concentrations of carbon monoxidemay be significantly different from ambient carbon monox-ide concentrations.

(2) Ambient outdoor concentrations of carbon monoxide withwhich people come in contact may vary significantly fromcarbon monoxide concentrations measured at fixed-sitemonitors.

In estimating exposure, the Council on Environmental Qualityalso assumed that each person in the population spends 24 h at home.


73

This assumption permitted the use of readily available demographicdata from the US Bureau of the Census. Data collected 20 years agoindicate that people spend a substantial portion of their time awayfrom home. In a study of metropolitan Washington, DC, residentsduring 1968, Chapin (1974) found that people spent an average 6.3h away from home on Sunday and 10.6 h away from home on Friday.This translates to between 26.3% and 44.2% of the day spent awayfrom home. More recent personal exposure and time budget studies(e.g., Johnson, 1987; Schwab et al., 1990) also indicate that a substan-tial portion of time is spent away from home.

Fixed-site monitors measure concentrations of pollutants inambient air. Ambient air has been defined by the US EPA in the Codeof Federal Regulations (OSHA, 1991b) as air that is “external tobuildings, to which the general public has access.” But the nature ofmodern urban lifestyles in many countries, including the USA,indicates that people spend an average of over 20 h per day indoors(Meyer, 1983). Reviews of studies on this subject by Yocom (1982),Meyer (1983) and Spengler & Soczek (1984) show that measurementsof indoor carbon monoxide concentrations vary significantly fromsimultaneous measurements in ambient air. The difference betweenindoor and outdoor air quality and the amount of time people spendindoors reinforce the conclusion that using ambient air quality mea-surements alone will not provide accurate estimates of populationexposure.

5.2.2 Approaches for estimating population exposure

In recent years, researchers have focused on the problem ofdetermining actual population exposures to carbon monoxide. Thereare three alternative approaches for estimating the exposures of apopulation to air pollution: the “direct approach,” using fieldmeasurement of a representative population carrying PEMs; the“indirect approach,” involving computation from field data of activitypatterns and measured concentration levels within microenvironments(Ott, 1982); and a hybrid approach that combines the direct andindirect approaches (Mage, 1991).

In the direct approach, as study participants engage in regulardaily activities, they are responsible for recording their exposures tothe pollutant of interest using a personal monitor. Subjects can recordtheir exposures in a diary, the method used in a US pilot study in


74

Los Angeles, California (Ziskind et al., 1982), or they can auto-matically store exposure data in a data logger, the method used instudies in Denver, Colorado (Johnson, 1984), and Washington, DC(Hartwell et al., 1984), which are summarized by Akland et al.(1985). In all of these studies, subjects also recorded the time andnature of their activities while they monitored personal exposures tocarbon monoxide.

The direct approach can be used to obtain an exposure inventoryof a representative sample from either the general population or aspecific subpopulation, which can be defined by many demographic,occupational and health factors. The inventory can cover a range ofmicroenvironments encountered over a period of interest (e.g., a day),or it can focus on one particular microenvironment. With this flexi-bility, policy analysts can assess the problem that emission sourcespose to a particular subgroup (e.g., commuters) active in a specificmicroenvironment (e.g., automobiles).

The indirect approach to estimating personal exposure is to usePEMs or microenvironmental monitors to monitor microenvironmentsrather than individuals. Combined with the ambient data andadditional data on human activities that occur in these microenviron-ments, data from the indirect approach can be used to estimate thepercentage of a subpopulation that is at risk for exposure to pollutantconcentrations that exceed national or regional air quality standards.Flachsbart & Brown (1989) conducted this type of study to estimatemerchant exposure to carbon monoxide from motor vehicle exhaustat the Ala Moana Shopping Center in Honolulu, Hawaii, USA.

5.2.3 Personal monitoring field studies

The development of small PEMs made possible the large-scalecarbon monoxide human exposure field studies in Denver, Colorado,and Washington, DC, in the winter of 1982–83 (Akland et al., 1985).These monitors proved effective in generating 24-h carbon monoxideexposure profiles on 450 persons in Denver and 800 persons inWashington, DC. The Denver–Washington, DC, study is the onlylarge-scale field study on population exposure to carbon monoxidethat has been undertaken to date.

Results from the Denver–Washington, DC, study (Akland et al.,1985) show that over 10% of the Denver residents and 4% of the


75

Washington, DC, residents were exposed to 8-h average carbonmonoxide levels above 10 mg/m3 (9 ppm) during the winter studyperiod. This degree of population exposure could not be accuratelydeduced from simultaneous data collected by the fixed-site monitorswithout taking into account other factors, such as contributions fromindoor sources, elevated levels within vehicles and individuals’activity patterns. In Denver, for example, the fixed-site monitorsexceeded the 10 mg/m3 (9 ppm) level only 3.1% of the time. Theseresults indicate that the effects of personal activity, indoor sourcesand, especially, time spent commuting all greatly contribute to aperson’s carbon monoxide exposure.

This study emphasizes that additional strategies are required toaugment data from fixed-site monitoring networks in order to evaluateactual human carbon monoxide exposures and health risks within acommunity. The cumulative carbon monoxide data for both Denverand Washington, DC, show that personal monitors often measurehigher concentrations than do fixed stations. As part of this study, 1-hexposures to carbon monoxide concentrations as determined bypersonal monitors were compared with measured ambient concentra-tions at fixed monitor sites. Correlations between personal monitordata and fixed-site data were consistently poor; the fixed-site datausually explained less than 10% of the observed variation in personalexposure. For example, 1-h carbon monoxide measurements taken atthe nearest fixed stations were only weakly correlated (0.14 # r #0.27) with office or residential measurements taken with personalmonitors (Akland et al., 1985).

The conclusion that exposure of persons to ambient carbon mon-oxide and other pollutants does not directly correlate with concentra-tions determined at fixed-site monitors is supported by the work ofothers (Ott & Eliassen, 1973; Cortese & Spengler, 1976; Dockery &Spengler, 1981; Wallace & Ott, 1982; Wallace & Ziegenfus, 1985).Results from the Finnish Liila study in Helsinki, in which personalcarbon monoxide and nitrogen dioxide exposures of preschool chil-dren were monitored, showed that their short-term personal carbonmonoxide exposures did not correlate with carbon monoxide levels inambient air and that gas stove use at home was the dominant deter-minant of carbon monoxide exposure (Alm et al., 1994).


76

In view of the high degree of variability of ambient carbonmonoxide concentrations over both space and time and the presenceof indoor sources of carbon monoxide, the reported results are notsurprising. A given fixed monitor is unable to track the exposure ofindividuals to ambient carbon monoxide as they go about their dailyactivities, moving from one location to another, all of which areseldom in the immediate vicinity of the monitor. This does notnecessarily mean, however, that fixed monitors do not give someuseful general information on the overall level of exposure of apopulation to carbon monoxide. The Denver and Washington, DC,data, although failing to show a correlation between exposuresmeasured by individual personal monitors and simultaneousconcentrations measured by the nearest fixed-site monitors, didsuggest that, in Denver, aggregate personal exposures were lower ondays of lower ambient carbon monoxide levels as determined by fixed-site monitors and higher on days of higher ambient levels. Also, bothfixed-site and personal exposures were higher in Denver than inWashington, DC. For example, the median ambient daily 1-h maxi-mum carbon monoxide concentration was measured by fixed monitorsto be 3.7 mg/m3 (3.2 ppm) higher in Denver than in Washington, DC,and the personal median daily 1-h maximum carbon monoxide expo-sure was measured by PEMs to be 4.5 mg/m3 (3.9 ppm) higher inDenver. Likewise, the median ambient daily 8-h maximum carbonmonoxide concentration measured by fixed monitors was found to be3.3 mg/m3 (2.9 ppm) higher in Denver, whereas the personal mediandaily 8-h maximum carbon monoxide exposure was 3.9 mg/m3

(3.4 ppm) higher in Denver.

The in-transit microenvironment with the highest estimatedcarbon monoxide concentration was the motorcycle, whereas walkingand bicycling had the lowest carbon monoxide concentrations. Out-door microenvironments can also be ranked for these data. Outdoorpublic garages and outdoor residential garages and carports had thehighest carbon monoxide concentrations; outdoor service stations,vehicle repair facilities and parking lots had intermediate concentra-tions. In contrast, school grounds and residential grounds hadrelatively low concentrations, whereas extremely low carbonmonoxide concentrations were found in outdoor sports arenas,amphitheatres, parks and golf courses. Finally, a wide range ofconcentrations was found in Denver within indoormicroenvironments. The highest indoor carbon monoxide


77

concentrations occurred in service stations, vehicle repair facilitiesand public parking garages; intermediate concentrations were foundin shopping malls, residential garages, restaurants, offices,auditoriums, sports arenas, concert halls and stores; and the lowestconcentrations were found in health care facilities, public buildings,manufacturing facilities, homes, schools and churches.

One activity that influences personal exposure is commuting. Anestimated 1% of the non-commuters in Washington, DC, wereexposed to concentrations above 10 mg/m3 (9 ppm) for 8 h. Bycomparison, an estimated 8% of persons reporting that theycommuted more than 16 h per week had 8-h carbon monoxideexposures above the 10 mg/m3 (9 ppm) level. Finally, certainoccupational groups whose work brings them in close proximity to theinternal combustion engine had a potential for elevated carbonmonoxide exposures. These include automobile mechanics; parkinggarage or gas station attendants; crane deck operators; cooks; taxi, busand truck drivers; firemen; policemen; and warehouse andconstruction workers. Of the 712 carbon monoxide exposure profilesobtained in Washington, DC, 29 persons fell into this “high-exposure” category. Of these, 25% had 8-h carbon monoxideexposures above the 10 mg/m3 (9 ppm) level.

Several field studies have also been conducted by the US EPA todetermine the feasibility and effectiveness of monitoring selectedmicroenvironments for use in estimating exposure profiles indirectly.One study (Flachsbart et al., 1987), conducted in Washington, DC, in1982 and 1983, concentrated on the commuting microenvironment,because earlier studies identified this microenvironment type as thesingle most important non-occupational microenvironment relative tototal carbon monoxide population exposure. It was observed that forthe typical automobile commuter, the time-weighted average carbonmonoxide exposure while commuting ranged from 10 to 16 mg/m3

(9 to 14 ppm). The corresponding rush-hour (7:00 to 9:00 a.m., 4:00to 6:00 p.m.) averages at fixed-site monitors were 3.1–3.5 mg/m3

(2.7–3.1 ppm).

5.2.4 Carbon monoxide exposures indoors

People in developed countries spend a majority (~85%) of theirtime indoors (US EPA, 1989a,b); therefore, a comprehensivedepiction of exposure to carbon monoxide must include this setting.The indoor sources, emissions and concentrations are sufficientlydiverse, however, that only a few studies can be cited here as


78

examples. Targeted field studies that have monitored indoor carbonmonoxide levels as a function of the presence or absence ofcombustion sources are described in more detail in section 5.6.

Early studies date back to before 1970, when it was found thatindoor and outdoor carbon monoxide levels do not necessarily agree.For example, one study determined indoor–outdoor relationships forcarbon monoxide over 2-week periods during summer, winter and fallin 1969 and 1970 in buildings in Hartford, Connecticut, USA (Yocomet al., 1971). With the exception of the private homes, which wereessentially equal, there was a day-to-night effect in the fall and winterseasons; days were higher by about a factor of 2. These differences areconsistent with higher traffic-related carbon monoxide levels outdoorsin the daytime.

Indoor and outdoor carbon monoxide concentrations weremeasured in four homes, also in the Hartford, Connecticut, area, in1973 and 1974 (Wade et al., 1975). All used gas-fired cooking stoves.Concentrations were measured in the kitchen, living room and bed-room. Stove use, as determined by activity diaries, correlated directlywith carbon monoxide concentrations. Peak carbon monoxide concen-trations in several of the kitchens exceeded 10 mg/m3 (9 ppm), butaverage concentrations ranged from 2.3–3.4 mg/m3 (2–3 ppm) toabout 9 mg/m3 (8 ppm). These results are in general agreement withresults obtained in Boston, Massachusetts, USA (Moschandreas &Zabransky, 1982). In this study, the investigators found significantdifferences between carbon monoxide concentrations in rooms inhomes where there were gas appliances.

Effects of portable kerosene-fired space heaters on indoor airquality were measured in an environmental chamber and a house(Traynor et al., 1982). Carbon monoxide emissions from white flameand blue flame heaters were compared. The white flame convectiveheater emitted less carbon monoxide than the blue flame radiantheater. Concentrations in the residence were <2.3 mg/m3 (<2 ppm)and 2.3–8 mg/m3 (2–7 ppm), respectively. The authors concluded thathigh levels might occur when kerosene heaters are used in smallspaces or when air exchange rates are low.

A rapid method using an electrochemical PEM to survey carbonmonoxide was applied in nine high-rise buildings in the San


79

Francisco and Los Angeles, California, areas during 1980 and 1984(Flachsbart & Ott, 1986). One building had exceptionally high carbonmonoxide levels compared with the other buildings; averageconcentrations on various floors ranged from 6 to 41 mg/m3 (5 to36 ppm). The highest levels were in the underground parking garage,which was found to be the source of elevated carbon monoxide withinthe building through transport via the elevator shaft.

The effect of residential wood combustion and specific heatertype on indoor carbon monoxide levels has been investigated(Humphreys et al., 1986). Airtight and non-airtight heaters werecompared in a research home in Tennessee, USA. Carbon monoxideemissions from the non-airtight heaters were generally higher thanthose from the airtight heaters. Peak indoor carbon monoxide concen-tration (ranging from 1.5 to 33.9 mg/m3 [1.3 to 29.6 ppm], dependingon heater type) was related to fuel reloadings.

Two studies in the Netherlands have measured carbon monoxidelevels in homes. Carbon monoxide levels in 254 Netherland homeswith unvented gas-fired geysers (water heaters) were investigatedduring the winter of 1980 (Brunekreef et al., 1982). Concentrationsat breathing height were grouped into the following categories:#11 mg/m3 (#10 ppm; n = 154), 13–57 mg/m3 (11–50 ppm; n = 50),58–110 mg/m3 (51–100 ppm; n = 25) and >110 mg/m3 (>100 ppm;n = 17). They found that a heater vent reduced indoor carbonmonoxide concentrations and that the type of burner affected carbonmonoxide levels. In another study, air pollution in Dutch homes wasinvestigated by Lebret (1985). Carbon monoxide concentrations weremeasured in the kitchen (0–20.0 mg/m3 [0–17.5 ppm]), the livingroom (0–10.0 mg/m3 [0–8.7 ppm]) and the bedroom (0–4.0 mg/m3

[0–3.5 ppm]). Carbon monoxide levels were elevated in homes withgas cookers and unvented geysers. Kitchen carbon monoxide levelswere higher than those in other locations as a result of peaks from theuse of gas appliances. Carbon monoxide levels in living rooms wereslightly higher in houses with smokers. The overall mean carbonmonoxide level indoors was 0–3.1 mg/m3 (0–2.7 ppm) above outdoorlevels.

In Zagreb, Yugoslavia, carbon monoxide was measured in eighturban institutions housing sensitive populations, including kinder-gartens, a children’s hospital and homes for the elderly (Sisovic &


80

Fugas, 1985). Winter carbon monoxide concentrations ranged from1.3 to 15.7 mg/m3 (1.1 to 13.7 ppm), and summer concentrationsranged from 0.7 to 7.9 mg/m3 (0.6 to 6.9 ppm). The authors attributedindoor carbon monoxide concentrations to nearby traffic density,general urban pollution, seasonal differences and day-to-day weatherconditions. Indoor sources were not reported.

Toxic levels of carbon monoxide were also found in measure-ments at six ice skating rinks (Johnson et al., 1975b). This study wasprompted by the reporting of symptoms of headache and nauseaamong 15 children who patronized one of the rinks. Carbon monoxideconcentrations were found to be as high as 350 mg/m3 (304 ppm)during operation of a propane-powered ice-resurfacing machine.Depending on skating activity levels, the ice-resurfacing operationwas performed for 10 min every 1–2 h. Because this machine wasfound to be the main source of carbon monoxide, using catalyticconverters and properly tuning the engine greatly reduced emissionsof carbon monoxide and, hence, reduced carbon monoxideconcentrations. Similar findings have been reported by Spengler et al.(1978), Lévesque et al. (1990), Paulozzi et al. (1993) and Lee et al.(1994).

5.2.5 Carbon monoxide exposures inside vehicles

Studies of carbon monoxide concentrations inside automobileshave also been reported over the past decade.

Petersen & Sabersky (1975) measured pollutants inside an auto-mobile under typical driving conditions. Carbon monoxide concentra-tions were generally less than 29 mg/m3 (25 ppm), with one 3-minpeak of 52 mg/m3 (45 ppm). Average concentrations inside the vehiclewere similar to those outside. No in-vehicle carbon monoxide sourceswere noted; however, a commuter’s exposure is usually determined byother high-emitting vehicles, not by the driven vehicle itself (Chan etal., 1989; Shikiya et al., 1989).

Drowsiness, headache and nausea were reported by eightchildren who had ridden in school buses for about 2 h while travellingon a ski trip (Johnson et al., 1975a). The students reporting symptomswere seated in the rear of the bus, which had a rear-mounted engineand a leaky exhaust. The exhaust system was subsequently repaired.During a later ski outing for students, carbon monoxideconcentrations were also monitored for a group of 66 school buses in


81

the parking lot. The investigators found 5 buses with carbonmonoxide concentrations of 6–29 mg/m3 (5–25 ppm) (mean 17 mg/m3

[15 ppm]), 24 buses showing concentrations in excess of 10 mg/m3

(9 ppm) for short periods and 2 buses showing up to 3 times the 10mg/m3 (9 ppm) level for short periods. Drivers were advised to parkso that exhausts from one bus would not be adjacent to the fresh airintake for another bus.

During a US cross-country trip in the spring of 1977, Chaney(1978) measured in-vehicle carbon monoxide concentrations. Thecarbon monoxide levels varied depending on traffic speed. Onexpressways in Chicago, Illinois, San Diego, California, and LosAngeles, California, when traffic speed was less than 16 km/h, carbonmonoxide concentrations exceeded 17 mg/m3 (15 ppm). Levelsincreased to 52 mg/m3 (45 ppm) when traffic stopped. In addition, itwas observed that heavily loaded vehicles (e.g., trucks) produced highcarbon monoxide concentrations inside nearby vehicles, especiallywhen the trucks were ascending a grade.

Colwill & Hickman (1980) measured carbon monoxide concen-trations in 11 new cars as they were driven on a heavily traffickedroute in and around London, United Kingdom. The inside meancarbon monoxide level for the 11 cars was 28.9 mg/m3 (25.2 ppm),whereas the outside mean level was 53.8 mg/m3 (47.0 ppm).

In a study mandated by the US Congress in the 1977 Clean AirAct Amendments, the EPA studied carbon monoxide intrusion intovehicles (Ziskind et al., 1981). The objective was to determinewhether carbon monoxide was leaking into the passenger compart-ments of school buses, police cars and taxis and, if so, how prevalentthe situation was. The study involved 1164 vehicles in Boston,Massachusetts, and Denver, Colorado. All vehicles were in use in aworking fleet at the time of testing. The results indicated that all threetypes of vehicles often have multiple (an average of four to five) pointsof carbon monoxide intrusion — worn gaskets, accelerator pedals,rust spots in the trunk, etc. In 58% of the rides lasting longer than 8h, carbon monoxide levels exceeded 10 mg/m3 (9 ppm). Thus, thestudy provided evidence that maintenance and possibly design ofvehicles may be important factors in human exposure to carbonmonoxide.


82

Petersen & Allen (1982) reported the results of carbon monoxidemeasurements taken inside vehicles under typical driving conditionsin Los Angeles, California, over 5 days in October 1979. They foundthat the average ratio of interior to exterior carbon monoxideconcentrations was 0.92. However, the hourly average interior carbonmonoxide concentrations were 3.9 times higher than the nearestfixed-site measurements. In their analysis of the factors that influenceinterior carbon monoxide levels, they observed that traffic flow andtraffic congestion (stop-and-go) are important, but “comfort state”(i.e., car windows open/closed, fan on/off, etc.) and meteorologicalparameters (i.e., wind speed, wind direction) have little influence onincremental exposures. Another study, carried out in Paris, France(Dor et al., 1995), also showed that the carbon monoxide concentra-tion in cars can be 3 times that of the ambient air. A study in HongKong showed that the same may be true for concentrations in buses(Chan & Wu, 1993).

Flachsbart (1989) investigated the effectiveness of priority laneson a Honolulu, Hawaii, arterial highway in reducing commuter traveltime and exposure to carbon monoxide. The carbon monoxideconcentrations and exposures of commuters in these lanes weresubstantially lower than in the non-priority lanes. Carbon monoxideexposure was reduced approximately 61% for express buses, 28% forhigh-occupancy vehicles and 18% for carpools when compared withthat for regular automobiles. The higher speed associated with prioritylanes helped reduce carbon monoxide exposure. These observationsdemonstrate that carbon monoxide concentrations have a high degreeof spatial variability on roadways, associated with vehicle speed andtraffic volume.

Ott et al. (1994) measured carbon monoxide exposures inside acar travelling on a major urban arterial highway in El Camino Realin the USA (traffic volume 30 500–45 000 vehicles per day) over a13.5-month period. For 88 trips, the mean carbon monoxideconcentration was 11.2 mg/m3 (9.8 ppm), with a standard deviationof 6.6 mg/m3 (5.8 ppm).

Fernadenz-Bremauntz & Ashmore (1995a,b) related exposure ofcommuters in vehicles to carbon monoxide to fixed-siteconcentrations at monitoring stations in Mexico City. The ambientlevels were all more than 15 mg/m3 (13 ppm). The highest median


83

and 90th percentile in-vehicle concentrations were found in autos andminibuses (49–68 and 77–96 mg/m3 [43–59 and 67–84 ppm]), withlower levels in buses (34 and 49 mg/m3 [30 and 43 ppm]), trolleys (30and 44 mg/m3 [26 and 38 ppm]) and metro and light rail (24 and 33mg/m3 [21 and 29 ppm]). Average in-vehicle/ambient ratios for eachmode of transport were as follows: automobile, 5.2; minivan, 5.2;minibus, 4.3; bus, 3.1; trolleybus, 3.0; and metro, 2.2.

Chan et al. (1991) investigated driver exposure to volatileorganic chemicals, carbon monoxide, ozone and nitrogen dioxideunder different urban, rural and interstate highway driving conditionsusing four different cars in Raleigh, North Carolina, USA. They foundthat the in-vehicle carbon monoxide concentrations did not vary signi-ficantly between the cars, and they were on average 4.5 times higherthan the ambient carbon monoxide measurements. Car ventilation hadlittle effect on the driver exposures.

Two Finnish studies of personal air pollution exposures ofchildren showed that preschool children who commuted to a day carecentre by bus or car were exposed to considerably higher peak carbonmonoxide levels than children who went to the day care centre bywalking or on a bike (Jantunen et al., 1995) and that the averagecarbon monoxide exposure of schoolchildren in Kuopio, Finland, ina car or bus was 4 times higher than in other microenvironments(Alm et al., 1995).

5.2.6 Carbon monoxide exposures outdoors

Carbon monoxide concentrations in outdoor settings (besidesthose measured at fixed monitoring stations) show considerablevariability. Ott (1971) made 1128 carbon monoxide measurements atoutdoor locations in San Jose, California, USA, at breathing heightover a 6-month period and compared these results with the officialfixed monitoring station data. This study simulated the measurementsof the outdoor carbon monoxide exposures of pedestrians in down-town San Jose by having them carry personal monitoring pumps andsampling bags while walking standardized routes on congested side-walks. If an outdoor measurement was made more than 100 m awayfrom any major street, its carbon monoxide concentration was similar,suggesting the existence of a generalized urban backgroundconcentration in San Jose that was spatially uniform over the city(within a 33-km2 grid) when one is sufficiently far away from mobile


84

sources. Because the San Jose monitoring station was located near astreet with heavy traffic, it recorded concentrations approximately100% higher than this background value. In contrast, outdoor carbonmonoxide levels from personal monitoring studies of downtownpedestrians were 60% above the corresponding monitoring stationvalues, and the correlation coefficient was low (r = 0.20). Bycollecting the pedestrian personal exposures over 8-h periods, it waspossible to compare the levels with the air quality standards. On 2 of7 days for which data were available, the pedestrian concentrationswere particularly high (15 and 16.3 mg/m3 [13 and 14.2 ppm]) andwere 2–3 times the corresponding levels recorded at the same time(5.0 and 7.1 mg/m3 [4.4 and 6.2 ppm]) at the air monitoring station(Ott & Eliassen, 1973; Ott & Mage, 1975). These results show thatconcentrations to which pedestrians were exposed on downtownstreets could exceed recommended air quality standards, although theofficial air monitoring station record values were significantly lessthan that. It can be argued, however, that not many pedestrians spenda lot of time outdoors walking along downtown sidewalks, and that isone of the important reasons for including realistic human activitypatterns in exposure assessments. It should also be noted that street-level carbon monoxide concentrations have significantly decreasedover the last decade in countries where vehicular emission controlsare in place.

Godin et al. (1972) conducted similar studies in downtownToronto, Ontario, Canada, using 100-ml glass syringes in conjunctionwith NDIR spectrometry. They measured carbon monoxide concentra-tions along streets, inside passenger vehicles and at a variety of otherlocations. Like other investigators, they found that carbon monoxideconcentrations were determined by very localized phenomena. Ingeneral, carbon monoxide concentrations in traffic and along streetswere much higher than those observed at conventional fixed air moni-toring stations. In a subsequent study in Toronto, Wright et al. (1975)used Ecolyzers to measure 4- to 6-min average carbon monoxideconcentrations encountered by pedestrians and street workers andobtained similar results. Levels ranged from 11 to 57 mg/m3 (10 to50 ppm), varying with wind speed and direction, atmosphericstability, traffic density and height of buildings. They also measuredcarbon monoxide concentrations on the sidewalks of a street thatsubsequently was closed to traffic to become a pedestrian mall. Beforethe street was closed, the average concentrations at two intersectionswere 10.8 ± 4.6 mg/m3 (9.4 ± 4.0 ppm) and 9.0 ± 2.2 mg/m3 (7.9 ±


85

1.9 ppm) (mean, plus or minus standard deviation); after the streetwas closed, the averages dropped to 4.2 ± 0.6 mg/m3 (3.7 ± 0.5 ppm)and 4.6 ± 1.1 mg/m3 (4.0 ± 1.0 ppm), respectively, which wereequivalent to the background level.

A large-scale field investigation was undertaken of carbon mon-oxide concentrations in indoor and outdoor locations in five Califor-nia, USA, cities using personal monitors (Ott & Flachsbart, 1982).For outdoor commercial settings, the average carbon monoxideconcentration was 5 mg/m3 (4 ppm). This carbon monoxide level wasstatistically, but not substantially, greater than the average carbonmonoxide concentration of 2.3 mg/m3 (2 ppm) recordedsimultaneously at nearby fixed monitoring stations.

The Organisation for Economic Co-operation and Development(OECD, 1997) recently published a study of trends and relative con-centrations in Western Europe, the USA and Japan between 1988 and1993, based on the annual maximum 8-h average concentration aturban traffic and urban residential sites. This showed that the averagelevels remained unchanged, but, at the most polluted locations in bothresidential and commercial (heavily trafficked) areas, levels of carbonmonoxide have declined significantly. A short summary of results ofdifferent countries is given in Table 6, as published by the OECD(1997). Carbon monoxide measurements in atmospheric air wereperformed in many countries — for example, at approximately 300sites in Germany using automatic devices.

Investigations in traffic routes of Amsterdam, Netherlands (vanWijnen et al., 1995), using personal air sampling resulted in muchhigher carbon monoxide concentrations in the samples of car driversthan in the personal air samples of cyclists. Similar results from Ger-many were cited.

5.3 Estimating population exposure to carbon monoxide

Accurate estimates of human exposure to carbon monoxide area prerequisite for both a realistic appraisal of the risks posed by thepollutant and the design and implementation of effective controlstrategies. This section discusses the general concepts on which expo-sure assessment is based and approaches for estimating populationexposure to carbon monoxide using exposure models. Because

Table 6. Comparison of carbon monoxide concentrations, 1993

Site Averagingperiod

Statistic Carbon monoxide concentration

Western Europe USA Japan

mg/m3 ppm mg/m3 ppm mg/m3 ppm

Urbantraffic

8-haverage

Average 8.3 7.2 6.8 5.9 5.5 4.8

8-haverage

95thpercentile

13.1 11.4 11.6 10.1 9.1 7.9

Urbanresidential

8-haverage

Average 5.7 5 6.4 5.6 4.3 3.8

8-haverage

95thpercentile

9.5 8.3 10.1 8.8 7.1 6.2


87

of problems in estimating population exposure solely from fixed-station data, several formal human exposure models have beendeveloped. Some of these models include information on humanactivity patterns: the microenvironments people visit and the timesthey spend there. These models also contain submodels depicting thesources and concentrations likely to be found in eachmicroenvironment, including indoor, outdoor and in-transit settings.

5.3.1 Components of exposure

Two aspects of exposure bear directly on the related healthconsequences. The first is the magnitude of the pollutant exposure.The second is the duration of the exposure. The magnitude is animportant exposure parameter, because concentration typically isassumed to be directly proportional to dose and, ultimately, to thehealth outcome. But exposure implies a time component, and it isessential to specify the duration of an exposure. The health risks ofexposure to a specific concentration for 5 min are likely to bedifferent, all other factors being equal, from those of exposure to thesame concentration for an hour.

The magnitude and duration of exposure can be determined byplotting an individual’s air pollution exposure over time. The functionCi(t) describes the air pollutant concentration to which an individualis exposed at any point in time t. Ott (1982) defined the quantity Ci(t)as the instantaneous exposure of an individual. The shaded area underthe graph represents the accumulation of instantaneous exposures oversome period of time (t1 ! t0). This area is also equal to the integral ofthe air pollutant concentration function, Ci(t), between t0 and t1. Ott(1982) defined the quantity represented by this area as the integratedexposure.

The average exposure, calculated by dividing the integratedexposure by the period of integration (t1 ! t0), represents the averageair pollutant concentration to which an individual was exposed overthe defined period of exposure. To facilitate comparison with estab-lished air quality standards, an averaging period is usually chosen toequal the averaging period of the standard. In this case, the averageexposure is referred to as a standardized exposure.

As discussed above, exposure represents the joint occurrence ofan individual being located at point (x,y,z) during time t with the


88

simultaneous presence of an air pollutant at concentration Cxyz(t).Consequently, an individual’s exposure to an air pollutant is afunction of location as well as time. If a volume at a location can bedefined such that air pollutant concentrations within it are relativelyhomogeneous, yet potentially different from other locations, thevolume may be considered a “microenvironment” (Duan, 1982).Microenvironments may be aggregated by location (e.g., indoor oroutdoor) or activity performed at a location (e.g., residential orcommercial) to form microenvironment types.

It is important to distinguish between individual exposures andpopulation exposures. Sexton & Ryan (1988) defined the pollutantconcentrations experienced by a specific individual during normaldaily activities as “personal” or “individual” exposures. A personalexposure depends on the air pollutant concentrations that are presentin the locations through which the person moves, as well as on thetime spent at each location. Because time–activity patterns can varysubstantially from person to person, individual exposures exhibit widevariability (Dockery & Spengler, 1981; Quackenboss et al., 1982;Sexton et al., 1984; Spengler et al., 1985; Stock et al., 1985; Wallaceet al., 1985). Thus, although it is a relatively straightforwardprocedure to measure any one person’s exposure, many suchmeasurements may be needed to quantify the mean and variance ofexposures for a defined group. The daily activities of a person in timeand space define his or her activity pattern. Accurate estimates of airpollution exposure generally require that an exposure model accountfor the activity patterns of the population of interest. The activitypatterns may be determined through “time budget” studies of thepopulation. Studies of this type have been performed by Szalai (1972),Chapin (1974), Michelson & Reed (1975), Robinson (1977), Johnson(1987) and Schwab et al. (1990). The earlier studies may now be outof date because they were not designed to investigate human exposurequestions and because lifestyles have changed over the past 25 years.Ongoing exposure studies have adopted the diary methods that weredeveloped for sociological investigations and applied them to currentexposure and time budget investigations. A few of these studies havebeen reported (e.g., Johnson, 1987; Schwab et al., 1990).

From a public health perspective, it is important to determine the“population exposure,” which is the aggregate exposure for a specifiedgroup of people (e.g., a community or an identified occupational


89

cohort). Because exposures are likely to vary substantially betweenindividuals, specification of the distribution of personal exposureswithin a population, including the average value and the associatedvariance, is often the focus of exposure assessment studies. The uppertail of the distribution, which represents those individuals exposed tothe highest concentrations, is frequently of special interest, becausethe determination of the number of individuals who experienceelevated pollutant levels can be critical for health risk assessments.This is especially true for pollutants for which the relationshipbetween dose and response is highly non-linear.

5.3.2 Approaches to exposure modelling

In recent years, the limitations of using fixed-site monitors aloneto estimate public exposure to air pollutants have stimulated interestin using portable monitors to measure personal exposure. These PEMswere developed for carbon monoxide in the late 1970s by EnergeticsScience Incorporated and by General Electric. Wallace & Ott (1982)surveyed PEMs available then for carbon monoxide and other airpollutants.

The availability of these monitors has facilitated use of the directand indirect approaches to assessing personal exposure (see section5.2.2). Whether the direct or indirect approach is followed, theestimation of population exposure requires a “model” — that is, amathematical or computerized approach of some kind. Sexton & Ryan(1988) suggested that most exposure models can be classified as oneof three types: statistical, physical or physical-stochastic.

The statistical approach requires the collection of data on humanexposures and the factors thought to be determinants of exposure.These data are combined in a statistical model, normally a regressionequation or an analysis of variance, to investigate the relationshipbetween air pollution exposure (dependent variable) and the factorscontributing to the measured exposure (independent variables). Anexample of a statistical model is the regression model developed byJohnson et al. (1986) for estimating carbon monoxide exposures inDenver, Colorado, based on data obtained from the Denver PersonalMonitoring Study. If the study group constitutes a representativesample, the derived statistical model may be extrapolated to thepopulation defined by the sampling frame. It should also be noted thatselection of factors thought to influence exposure has a substantial


90

effect on the outcome of the analysis. Spurious conclusions can bedrawn, for example, from statistical models that include parametersthat are correlated with, but not causally related to, air pollutionexposure.

In the physical modelling approach, the investigator makes ana priori assumption about the underlying physical processes thatdetermine air pollution exposure and then attempts to approximatethese processes through a mathematical formulation. Because themodel is chosen by the investigator, it may produce biased resultsbecause of the inadvertent inclusion of inappropriate parameters orthe improper exclusion of critical components. The National AmbientAir Quality Standards (NAAQS) Exposure Model as originallyapplied to carbon monoxide by Johnson & Paul (1983) is an exampleof a physical model.

The physical-stochastic approach combines elements of both thephysical and statistical modelling approaches. The investigator beginsby constructing a mathematical model that describes the physicalbasis for air pollution exposure. Then, a random or stochasticcomponent that takes into account the imperfect knowledge of thephysical parameters that determine exposure is introduced into themodel. The physical-stochastic approach limits the effect ofinvestigator-induced bias by the inclusion of the random componentand allows for estimates of population distributions for air pollutionexposure. Misleading results may still be produced, however, becauseof poor selection of model parameters. In addition, the requiredknowledge about distributional characteristics may be difficult toobtain. Examples of models based on this approach that have beenapplied to carbon monoxide include the Simulation of Human Activityand Pollutant Exposure model (Ott, 1984; Ott et al., 1988) and twomodels derived from the NAAQS Exposure Model, developed byJohnson et al. (1990).

5.4 Exposure measurements in populations andsubpopulations

5.4.1 Carboxyhaemoglobin measurements in populations

Numerous studies have used the above-described methodologiesto characterize the levels of carboxyhaemoglobin in the general


91

population. These studies have been designed to determine frequencydistributions of carboxyhaemoglobin levels in the populations beingstudied. In general, the higher the frequency of carboxyhaemoglobinlevels above baseline in non-smoking subjects, the greater theincidence of significant carbon monoxide exposure.

Carboxyhaemoglobin levels in blood donors have been studiedfor various urban populations in the USA. Included have been studiesof blood donors and sources of carbon monoxide in the metropolitanSt. Louis, Missouri, population (Kahn et al., 1974); evaluation ofsmoking and carboxyhaemoglobin in the St. Louis metropolitanpopulation (Wallace et al., 1974); carboxyhaemoglobin analyses of16 649 blood samples provided by the Red Cross Missouri–Illinoisblood donor programme (Davis & Gantner, 1974); a survey of blooddonors for percent carboxyhaemoglobin in Chicago, Illinois,Milwaukee, Wisconsin, New York, New York, and Los Angeles,California (Stewart et al., 1976); a national survey for carboxy-haemoglobin in American blood donors from urban, suburban andrural communities across the USA (Stewart et al., 1974); and thetrend in percent carboxyhaemoglobin associated with vehicular trafficin Chicago blood donors (Stewart et al., 1976). These extensivestudies of volunteer blood donor populations show three main sourcesof exposure to carbon monoxide in urban environments: smoking,general activities (usually associated with internal combustionengines) and occupational exposures. For comparisons of sources, thepopulations are divided into two main groups — smokers and non-smokers. The main groups are often divided further into subgroupsconsisting of industrial workers, drivers, pedestrians and others, forexample. Among the two main groups, smokers show an average of4% carboxyhaemoglobin, with a usual range of 3–8%; non-smokersaverage about 1% carboxyhaemoglobin (Radford & Drizd, 1982).Smoking behaviour generally occurs as an intermittent diurnalpattern, but carboxyhaemoglobin levels can rise to a maximum ofabout 15% in some individuals who chain-smoke. Similar results wereobtained in a more recent study in Bahrain (Madany, 1992) and in astudy in Beijing, People’s Republic of China (Song et al., 1984).

Aside from tobacco smoke, the most significant sources ofpotential exposure to carbon monoxide in the population are commu-nity air pollution, occupational exposures and household exposures(Goldsmith, 1970). Community air pollution comes mainly from


92

automobile exhaust and has a typical intermittent diurnal pattern.Occupational exposures occur for up to 8 h per day, 5 days per week,producing carboxyhaemoglobin levels generally less than 10%.

More recent studies characterizing carboxyhaemoglobin levels inthe population have appeared in the literature. Turner et al. (1986)used an IL 182 CO-Oximeter to determine percent carboxyhaemo-globin in venous blood of a study group consisting of both smokingand non-smoking hospital staff, inpatients and outpatients. Bloodsamples were collected for 3487 subjects (1255 non-smokers) duringmorning hours over a 5-year period. A detailed smoking history wasobtained at the time of blood collection. Using 1.7% carboxyhaemo-globin as a normal cut-off value, the distribution for the populationstudied showed above-normal results for 94.7% of cigarette smokers,10.3% of primary cigar smokers, 97.4% of those exposed to environ-mental smoke from cigars and 94.7% of those exposed to environmen-tal smoke from pipes.

Zwart & van Kampen (1985) tested a blood supply using aroutine spectrophotometric method for total haemoglobin and for car-boxyhaemoglobin in 3022 samples of blood for transfusion in hospitalpatients in the Netherlands. For surgery patients over a 1-year period,the distribution of percent carboxyhaemoglobin in samples collectedas part of the surgical protocol showed 65% below 1.5% carboxy-haemoglobin, 26.5% between 1.5% and 5% carboxyhaemoglobin,6.7% between 5% and 10% carboxyhaemoglobin and 0.3% in excessof 10% carboxyhaemoglobin. This distribution of percent carboxy-haemoglobin was homogeneous across the entire blood supply,resulting in 1 in 12 patients having blood transfusions at 75%available haemoglobin capacity.

Radford & Drizd (1982) analysed blood carboxyhaemoglobin inapproximately 8400 samples obtained from respondents in the 65geographic areas of the second US National Health and NutritionExamination Survey during the period 1976–1980. When the fre-quency distributions of blood carboxyhaemoglobin levels are plottedon a logarithmic probability scale to facilitate comparison of theresults for different age groups and smoking habits, it is evident thatadult smokers in the USA have carboxyhaemoglobin levels consid-erably higher than those of non-smokers, with 79% of the smokers’blood samples above 2% carboxyhaemoglobin and 27% of the


93

observations above 5% carboxyhaemoglobin. The nationwide distribu-tions of persons aged 12–74 who have never smoked and who are ex-smokers were similar, with 5.8% of the ex-smokers and 6.4% of thenever-smokers above 2% carboxyhaemoglobin. It is evident that asignificant proportion of the non-smoking US population had bloodlevels above 2% carboxyhaemoglobin. For these two non-smokinggroups, blood levels above 5% were found in 0.7% of the never-smokers and 1.5% of the ex-smokers. It is possible that these highblood levels could be due, in part, to misclassification of somesmokers as either ex- or non-smokers. Children aged 2–11 had lowercarboxyhaemoglobin levels than the other groups, with only 2.3% ofthe children’s samples above 2% carboxyhaemoglobin and 0.2%above 5% carboxyhaemoglobin.

Wallace & Ziegenfus (1985) utilized available data from thesecond National Health and Nutrition Examination Survey to analysethe relationship between measured carboxyhaemoglobin levels and theassociated 8-h carbon monoxide concentrations at nearby fixed moni-tors. Carboxyhaemoglobin data were available for a total of 1658 non-smokers in 20 cities. The authors concluded that fixed outdoor carbonmonoxide monitors alone are, in general, not providing useful esti-mates of carbon monoxide exposure of urban residents.

5.4.2 Breath measurements in populations

In a study by Wallace (1983) in which breath measurements ofcarbon monoxide were used to detect an indoor air problem,65 workers in an office had been complaining for some months oflate-afternoon sleepiness and other symptoms, which they attributedto the new carpet. About 40 of the workers had their breath tested forcarbon monoxide on a Friday afternoon and again on a Mondaymorning. The average breath carbon monoxide levels decreased from26 mg/m3 (23 ppm) on Friday to 8 mg/m3 (7 ppm) on Mondaymorning, indicating a work-related condition. Non-working fans inthe parking garage and broken fire doors were identified as the causeof the problem.

Wald et al. (1981) obtained measurements of percent carboxy-haemoglobin for 11 749 men, aged 35–64, who attended a medicalcentre in London, United Kingdom, for comprehensive health screen-ing examinations between 11:00 a.m. and 5:00 p.m. The time ofsmoking for each cigarette, cigar or pipe smoked since waking was


94

recorded at the time of collection of a venous blood sample. Percentcarboxyhaemoglobin was determined using an IL 181 CO-Oximeter.Using 2% carboxyhaemoglobin as a normal cut-off value, 81% ofcigarette smokers, 35% of cigar and pipe smokers and 1% of non-smokers were found to be above normal. An investigation ofcarboxyhaemoglobin and alveolar carbon monoxide was conducted ona subgroup of 187 men (162 smokers and 25 non-smokers). Threesamples of alveolar air were collected at 2-min intervals within 5 minof collecting venous blood for carboxyhaemoglobin estimation.Alveolar air was collected by having the subject hold his breath for20 s and then exhale through a 1-m glass tube with an internaldiameter of 17 mm and fitted with a 3-litre anaesthetic bag at thedistal end. Air at the proximal end of the tube was considered to bealveolar air, and a sample was removed by a small side tube located5 mm from the mouthpiece. The carbon monoxide content wasmeasured using an Ecolyzer. The instrumental measurement is basedon detection of the oxidation of carbon monoxide to carbon dioxide bya catalytically active electrode in an aqueous electrolyte. The mean ofthe last two readings to the nearest 0.29 mg/m3 (0.25 ppm) wasrecorded as the alveolar carbon monoxide concentration. Subjectsreporting recent alcohol consumption were excluded, because ethanolin the breath affects the response of the Ecolyzer. A linear regressionequation of percent carboxyhaemoglobin on alveolar carbon monoxidelevel had a correlation coefficient of 0.97, indicating that acarboxyhaemoglobin level could be estimated reliably from analveolar carbon monoxide level.

Honigman et al. (1982) determined alveolar carbon monoxideconcentrations by end-expired breath analysis for athletes (joggers).The group included 36 non-smoking males and 7 non-smokingfemales, all conditioned joggers, covering at least 34 km per week forthe previous 6 months in the Denver, Colorado, area. The participantsexercised for a 40-min period each day over one of three definedcourses in the Denver urban environment (elevation 1610 m).Samples of expired air were collected and analysed before start ofexercise, after 20 min and again at the end of the 40-min exerciseperiod. Heart rate measurements at 20 min and 40 min were 84 and82% of mean age-predicted maxima, respectively, indicating exercisein the aerobic range. Relative changes in carbon monoxideconcentrations in expired air were plotted and compared with carbonmonoxide concentrations in ambient air measured at the time of


95

collecting breath samples. Air and breath samples were analysedusing an MSA Model 70. Relative changes in expired end-air carbonmonoxide based on the concentration of carbon monoxide in breathbefore the start of exercise were plotted in terms of the ambient airconcentrations measured during the exercise period, at both 20 and40 min of exercise. For ambient concentrations of carbon monoxidebelow 7 mg/m3 (6 ppm), the aerobic exercise served to decrease therelative amount of expired end-air carbon monoxide compared withthe concentration measured before the start of exercise. For ambientconcentrations in the range of 7–8 mg/m3 (6–7 ppm), there was no netchange in the carbon monoxide concentrations in the expired air. Forambient air concentrations in excess of 8 mg/m3 (7 ppm), the aerobicexercise resulted in relative increases in expired carbon monoxide,with the increases after 40 min being greater than similar increasesobserved at the 20-min measurements. Sedentary controls at themeasurement stations showed no relative changes. Thus, aerobicexercise, as predicted by the physiological models of uptake andelimination, is shown to enhance transport of carbon monoxide,thereby decreasing the time to reach equilibrium conditions.

Verhoeff et al. (1983) surveyed 15 identical residences that usednatural gas for cooking and geyser units for water heating. Carbonmonoxide concentrations in the flue gases were measured using anEcolyzer (2000 series). The flue gases were diluted to the dynamicrange of the instrument for carbon monoxide (determined by Draegertube analyses for carbon dioxide dilution to 2.0–2.5%). Breathsamples were collected from 29 inhabitants by having each participanthold a deep breath for 20 s and exhale completely through a glasssampling tube (225-ml volume). The sampling tube was stopperedand taken to a laboratory for analysis of carbon monoxide contentusing a gas–liquid chromatograph (Hewlett-Packard 5880A). Theoverall coefficient of variation for sampling and analysis was 7%,based on results of previous measurements. No significant differenceswere observed for non-smokers as a result of their cooking ordishwashing activities using the natural gas fixtures. There was aslight increase in carbon monoxide in expired air for smokers, but thismay be due to the possibility of increased smoking during the dinnerhour.

Wallace et al. (1984) reported data on measurements of expiredend-air carbon monoxide and comparisons with predicted valuesbased on personal carbon monoxide measurements for populations in


96

Denver, Colorado, and Washington, DC. Correlations between breathcarbon monoxide and preceding 8-h average carbon monoxide expo-sures were high (0.6–0.7) in both cities. Correlation coefficients werecalculated for 1-h to 10-h average personal carbon monoxideexposures in 1-h increments; the highest correlations occurred at7–9 h. However, breath carbon monoxide levels showed no relation-ship with ambient carbon monoxide measurements at the nearestfixed-station monitor.

A major large-scale study employing breath measurements ofcarbon monoxide was carried out by the US EPA in Washington, DC,and Denver, Colorado, in the winter of 1982–83 (Hartwell et al.,1984; Johnson, 1984; Wallace et al., 1984, 1988; Akland et al., 1985).In Washington, DC, 870 breath samples were collected from 812participants; 895 breath samples were collected from 454 Denverparticipants (two breath samples on 2 consecutive days in Denver).All participants also carried personal monitors to measure theirexposures over a 24-h period in Washington, DC, or a 48-h period inDenver. The subjects in each city formed a probability samplerepresenting 1.2 million adult non-smokers in Washington, DC, and500 000 adult non-smokers in Denver.

The distributions of breath levels in the two cities appeared to beroughly lognormal, with geometric means of 6.0 mg carbon monox-ide/m3 (5.2 ppm) for Denver and 5.0 mg carbon monoxide/m3

(4.4 ppm) for Washington, DC. Geometric standard deviations wereabout 1.8 mg/m3 (1.6 ppm) for each city. Arithmetic means were8.1 mg/m3 (7.1 ppm) for Denver and 6.0 mg/m3 (5.2 ppm) forWashington, DC.

Of greater regulatory significance is the number of people whosecarboxyhaemoglobin levels exceeded the value of 2.1%, because theUS EPA has determined that the current 10 mg/m3 (9 ppm), 8-haverage standard would keep more than 99.9% of the most sensitivenon-smoking adult population below this level of protection (FederalRegister, 1985). An alveolar carbon monoxide concentration of about11 mg/m3 (10 ppm) would correspond to a carboxyhaemoglobin levelof 2%. The percentage of people with measured breath valuesexceeding this level was about 6% in Washington, DC. This percen-tage was increased to 10% when the correction for the effect of roomair was applied. Of course, because the breath samples were taken on


97

days and at times when carbon monoxide levels were not necessarilyat their highest level during the year, these percentages are lowerlimits of the estimated number of people who may have incurredcarboxyhaemoglobin levels above 2%. Yet the two central stations inWashington, DC, recorded a total of one exceedance of the 10 mg/m3

(9 ppm) standard during the winter of 1982–83. Models based onfixed-station readings would have predicted that an exceedingly tinyproportion of the Washington, DC, population received exposuresexceeding the standard. Therefore, the results from the breath mea-surements indicated that a much larger portion of both Denver andWashington, DC, residents were exceeding 2% carboxyhaemoglobinthan was predicted by models based on fixed-station measurements.

It should also be noted that the number of people with measuredmaximum 8-h exposures exceeding the EPA outdoor standard of10 mg/m3 (9 ppm) was only about 3.5% of the Washington, DC,subjects. This value appears to disagree with the value of 10%obtained from the corrected breath samples. However, the personalmonitors used in the study were shown to experience several differentproblems, including a loss of response associated with battery dis-charge towards the end of the 24-h monitoring period, which causedthem to read low just at the time the breath samples were beingcollected. Therefore, Wallace et al. (1988) concluded that the breathmeasurements were correct and the personal air measurements werebiased low. The importance of including breath measurements infuture exposure and epidemiology studies is indicated by this study.

Hwang et al. (1984) described the use of expired air analysis forcarbon monoxide in an emergency clinical setting to diagnose thepresence and extent of carbon monoxide intoxication. The subjectswere 47 Korean patients brought in for emergency treatment whoshowed various levels of consciousness: alertness (11), drowsiness(21), stupor (7), semicoma (5), coma (1) and unknown (2). The studygroup included 16 males, aged 16–57, and 31 females, aged 11–62.Exposure durations ranged from 2 to 10 h, with all exposuresoccurring in the evening and nighttime hours. The source of carbonmonoxide was mainly charcoal fires used for cooking and heating. Inorder to estimate carbon monoxide concentrations in expired air, adetector tube (Gastec 1La containing potassium palladosulfite as botha reactant and colour-change indicator for the presence of carbonmonoxide on silica gel) was fitted to a Gastec manual sampling pump.


98

One stroke of the sampling plunger represents 100 ml of air. A 100-ml sample of expired air was collected by inserting a detector tube ata nostril and slowly pulling back the plunger for one full stroke forexpired air. A 10-ml sample of venous blood was also collected at thistime for determining percent carboxyhaemoglobin using a CO-Oximeter. The subjects showed signs of acute intoxication, and signi-ficant relationships were found between carbon monoxide levels inexpired air and percent carboxyhaemoglobin.

Cox & Whichelow (1985) analysed end-exhaled air (collectedover approximately the last half of the exhalation cycle) for carbonmonoxide concentrations for a random population of 168 adults —69 smokers and 99 non-smokers. The results were used to evaluate theinfluence of home heating systems on exposures to and absorption ofcarbon monoxide. Ambient indoor concentrations of carbon monoxidewere measured in the homes of study subjects. The subjects included86 men and 82 women, ranging in age from 18 to 74. Interviews wereusually conducted in the living room of the subject’s home. The typeof heating system in use was noted, and the indoor air concentrationof carbon monoxide was measured using an Ecolyzer. After theambient indoor carbon monoxide level was determined, a breathsample was collected from the subject. The subject was asked to holda deep breath for 20 s and then to exhale completely into a trilaminateplastic bag. The bag was fitted to the port of the Ecolyzer, and thecarbon monoxide content of the exhaled air was measured. Forsmokers, the time since smoking their last cigarette and the numberof cigarettes per day were noted. For non-smokers, there was a strongcorrelation between carbon monoxide levels in ambient air and carbonmonoxide levels in expired air. With smokers, the correlation wasstrongest with the number of cigarettes smoked per day. The data alsosupported the supposition that smokers are a further source of ambientcarbon monoxide in the indoor environment.

Lambert et al. (1988) compared carbon monoxide levels in breathwith carboxyhaemoglobin levels in blood in 28 subjects (including2 smokers). Breath carbon monoxide was collected using the standardtechnique developed by Jones et al. (1958): maximal inspiration wasfollowed by a 20-s breath-hold, and the first portion of the expiredbreath was discarded. Excellent precision (± 0.23 mg/m3 [± 0.2 ppm])was obtained in 35 duplicate samples. Blood samples were collectedwithin 15 min of the breath samples using a gas-tight plastic syringe


99

rinsed with sodium heparin. Carboxyhaemoglobin was measuredusing an IL 282 CO-Oximeter. Some samples were also measuredusing a gas chromatograph.

By using least squares regression, the null hypothesis of nodifference in the slope and intercept estimates for non-smokers andsmokers was not rejected (i.e., there was no association between bloodcarboxyhaemoglobin and breath carbon monoxide in either non-smokers or smokers).

5.4.3 Subject age

The relationship between age and carboxyhaemoglobin level isnot well established. Kahn et al. (1974) reported that non-smokingsubjects under the age of 19 years had a significantly lower percentcarboxyhaemoglobin than older subjects, but there was no differencein carboxyhaemoglobin between the ages of 20 and 59 years. Kahnet al. (1974) also reported that there was a slight decrease in thecarboxyhaemoglobin levels in non-smoking subjects over the age of60 years. Radford & Drizd (1982) reported that younger subjects,3–11 years old, had lower levels of carboxyhaemoglobin than did theolder age group of 12–74 years. Goldsmith (1970) reported thatexpired carbon monoxide levels were unchanged with age in non-smokers; however, there was a steady decline in the expired carbonmonoxide levels with age in smokers. The decrease in expired carbonmonoxide is disproportionately large for the decrease in carboxy-haemoglobin levels measured by Kahn et al. (1974) in older subjects.Therefore, by comparison of the data from these two studies, it wouldappear that older subjects have higher levels of carboxyhaemoglobinthan predicted from the expired carbon monoxide levels. It is notknown how much of this effect is due to aging of the pulmonarysystem, resulting in a condition similar to that of subjects withobstructive pulmonary disease (see below).

5.4.4 Pulmonary disease

A major potential influence on the relationship between bloodand alveolar partial pressures of carbon monoxide is the presence ofsignificant lung disease. Hackney et al. (1962) demonstrated that theslow increase in exhaled carbon monoxide concentration in arebreathing system peaked after 1.5 min in healthy subjects butrequired 4 min in a subject with lung disease. These findings have


100

been substantiated by Guyatt et al. (1988), who reported that patientswith pulmonary disease did not have the same relationship betweenpercent carboxyhaemoglobin and breath-hold carbon monoxide con-centrations. The group with pulmonary disease had a forcedexpiratory volume in 1 s (FEV1)/forced vital capacity (FVC) percent-age of <71.5%, whereas the healthy subjects had an FEV1/FVCpercentage of >86%. The linear regression for the healthy group wasCOHb = 0.629 + 0.158(ppm CO); for the pulmonary disease group,the linear regression was COHb = 0.369 + 0.185(ppm CO). Thismeans that at low carbon monoxide levels, individuals with obstruc-tive pulmonary disease would have a lower “alveolar” carbon mon-oxide level for any given percent carboxyhaemoglobin level thanwould the healthy subjects.

5.4.5 Effects of smoking

Studies evaluating the effect of cigarette smoking on end-expiredcarbon monoxide have found a phasic response that depends onsmoking behaviour (Henningfield et al., 1980; Woodman et al.,1987). There is an initial rapid increase in the carbon monoxideconcentration of expired air as a result of smoking. This is followedby a rapid (5-min) decrease after cessation of smoking and a slowdecrease over the 5- to 60-min period after smoking. A comparison ofthe results from one study (Tsukamoto & Matsuda, 1985) showed thatthe carbon monoxide concentration in expired air increases byapproximately 6 mg/m3 (5 ppm) after smoking one cigarette. Thiscorresponds to an increase of 0.67% carboxyhaemoglobin based onblood–breath relationships developed by the authors. Use of cigaretteswith different tar and nicotine yields or the use of filter-tip cigarettesshowed no apparent effect on end-expired carbon monoxideconcentrations (Castelli et al., 1982).

The relationship between breath-hold carbon monoxide andblood carbon monoxide is apparently altered as a result of smoking,making the detection of small changes difficult. Guyatt et al. (1988)showed that smoking one cigarette results in a variable response inthe relationship between breath-hold alveolar carbon monoxidefraction [FACO(Bh)] and carboxyhaemoglobin levels. The range ofFACO(Bh) values for an increase of 1% carboxyhaemoglobin was from!6 to +6 mg/m3 (!5 to +5 ppm). The correlation between the changein FACO(Bh) and the change in carboxyhaemoglobin in 500 subjectswas only 0.705. This r value indicates that only 50% of the change inFACO(Bh) was due to changes in carboxyhaemoglobin. It is not


101

known how much of this residual error is due to subject complianceor to error in the method. Therefore, the results obtained with breath-holding in smoking subjects should be viewed with caution unlesslarge differences in FACO(Bh) are reported (i.e., considerable cigaretteconsumption is being evaluated).

5.5 Megacities and other major urban areas

Throughout the world, there are large urban areas that haveserious air pollution problems and encompass large land areas andover 10 million people (total population of the 20 megacities in 1990was estimated to be 275 million). The cities that have been designatedas megacities include 2 in North America (Los Angeles, California,USA; New York, New York, USA), 4 in Central and South America(Buenos Aires, Argentina; Mexico City, Mexico; Rio de Janeiro,Brazil; São Paulo, Brazil), 1 in Africa (Cairo, Egypt), 11 in Asia(Bangkok, Thailand; Beijing, People’s Republic of China; Bombay,India; Calcutta, India; Delhi, India; Jakarta, Indonesia; Karachi,Pakistan; Manila, Philippines; Seoul, South Korea; Shanghai,People’s Republic of China; Tokyo, Japan) and 2 in Europe (London,United Kingdom; Moscow, Russia). However, many other cities areheading for megacity status.

Of the 20 megacities studied (WHO/UNEP, 1992; Mage et al.,1996), monitoring capabilities for carbon monoxide have been desig-nated as none or unknown in 9, rudimentary in 1, adequate in 4 andgood in 6 (see Table 7). The air quality for carbon monoxide has beendesignated as no data available or insufficient data for assessment insix; serious problems, WHO guidelines exceeded by more than afactor of 2 in one; moderate to heavy pollution, WHO guidelinesexceeded by up to a factor of 2 (short-term guidelines exceeded on aregular basis at certain locations) in seven; and low pollution, WHOguidelines are normally met (short-term guidelines may be exceededoccasionally) in six (see Table 7).

The problem of air pollution in Mexico City is confounded bygeography. The metropolitan area of Mexico City is located at a meanaltitude of 2240 m and is situated in the Mexican Basin. The reducedoxygen in the air at high altitude causes carbon monoxide emissionsto increase because of incomplete combustion, and it exacerbates thehealth effects attributed to carbon monoxide, especially among highly


102

Table 7. Overview of carbon monoxide air quality and monitoring capabilities in 20megacitiesa

Megacity Air quality for carbonmonoxideb

Status of monitoringcapabilitiesc

Bangkok, Thailand a B

Beijing, People’s Republicof China

d B

Bombay, India a D

Buenos Aires, Argentina d D

Cairo, Egypt b D

Calcutta, India d D

Delhi, India a D

Jakarta, Indonesia b D

Karachi, Pakistan d D

London, United Kingdom b B

Los Angeles, USA b A

Manila, Philippines d D

Mexico City, Mexico c A

Moscow, Russia b C

New York, USA b A

Rio de Janeiro, Brazil a B

São Paulo, Brazil b A

Seoul, South Korea a A

Shanghai, People’sRepublic of China

d D

Tokyo, Japan a A

a Adapted from WHO/UNEP (1992); Mage et al. (1996).b a = Low pollution, WHO guidelines normally met (short-term guidelines may be

exceeded occasionally). b = Moderate to heavy pollution, WHO guidelines exceeded by up to a factor of 2

(short-term guidelines exceeded on a regular basis at certain locations). c = Serious problem, WHO guidelines exceeded by more than a factor of 2. d = No data available or insufficient data for assessment.c A = Good. B = Adequate. C = Rudimentary. D = None or not known.


103

susceptible population groups, including children and pregnantwomen.

Data for air quality trends in about 40 countries were collectedin the WHO/United Nations Environment Programme (UNEP) GlobalEnvironmental Monitoring System (GEMS/Air) (UNEP/WHO, 1993).The GEMS/Air programme was terminated in 1995. WHO has set upa successive programme, the Air Management Information System(WHO, 1997a). Summary air quality data for carbon monoxide arecurrently available for many countries, such as Austria, Germany,Greece, Japan, New Zealand and Switzerland. For example, between1986 and 1995, the annual mean carbon monoxide concentrationsranged from 5.1 to 8.4 mg/m3 (4.5 to 7.3 ppm) in Greece, from 0.8 to2.7 mg/m3 (0.7 to 2.4 ppm) in Japan and from 0.9 to 1.9 mg/m3 (0.8to 1.7 ppm) in New Zealand. In Australia, the annual mean carbonmonoxide concentrations from 1988 to 1995 ranged from 0.66 to1.36 mg/m3 (0.58 to 1.19 ppm). Air quality trends for carbonmonoxide over the same period are shown in Fig. 3 for Athens,Greece, Chongqing, People’s Republic of China, Frankfurt, Germany,Johannesburg, South Africa, London, United Kingdom, and LosAngeles, USA.

In a recent report (Eerens et al., 1995), carbon monoxide datafrom 105 cities in 35 European states are given. They include annualaverage 8-h maximum concentrations and number of days exceedingWHO Air Quality Guidelines and twice the WHO Air Quality Guide-lines.

5.6 Indoor concentrations and exposures

Indoor concentrations of carbon monoxide are a function ofoutdoor concentrations, indoor sources (source type, source condition,source use, etc.), infiltration/ventilation and air mixing between andwithin rooms. In residences without sources, average carbonmonoxide concentrations are approximately equal to average outdoorconcentrations at the corresponding elevation, generally decreasingwith height above the ground. Proximity to outdoor sources (i.e.,structures near heavily travelled roadways or with attached garages orparking garages) can have a major impact on indoor carbon monoxideconcentrations.


104

Johannesburg, South Africa

Concentration[µg m-3]

Year

City centre

30003500

2500

200015001000

5000

19871986 1988 1989 1990 19951991 1992 1993 1994

London, United Kingdom


Year

Carbon monoxide annualmeans

5000

01986

Urban centre

Kerbside

4000300020001000

1987 1988 1989 1990 1991 1992 1993 1994 1995

Urban background

5000

Frankfurt, Germany


Year

Carbon Monoxide4000

3000

2000

1000

01986

City centre /commercialResidential

1987 1988 1989 1990 1991 1992 1993 1994 1995

Fig. 3. Trends in carbon monoxide air quality between 1986 and 1995 inJohannesburg (South Africa), Frankfurt (Germany), London (United Kingdom),

Athens (Greece), Los Angeles (USA) and Chongqing (China) (from WHO, 1997a).


105

10000

Athens, Greece


Year

Carbon Monoxide

8000

6000

4000

2000

01986 1987 1988 1989 1990 1991 1992 1993 1994 1995

Industrial

CommercialResidential

Los Angeles, USA

Concentration(ppm)

Year

95-percentile

15

10

5

0

19871986 1988 1989 1990 1991 1992 1993 1994 1995

Chongqing, China


Year

Urban

3000

1983

3500

25002000

15001000500

01984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994

Fig. 3. (contd).


106

The development of small lightweight and portable electrochemi-cal carbon monoxide monitors over the past decade has permitted themeasurement of personal carbon monoxide exposures and carbonmonoxide concentrations in a number of indoor environments. Theavailable data on indoor carbon monoxide concentrations have beenobtained from total personal exposure studies or studies in whichvarious indoor environments have been targeted for measurements.

The extensive total personal carbon monoxide exposure studiesconducted by the US EPA in Washington, DC, and Denver, Colorado,have shown that the highest carbon monoxide concentrations occur inindoor microenvironments associated with transportation sources(parking garages, cars, buses, etc.). Concentrations in these environ-ments were found to frequently exceed 10 mg/m3 (9 ppm). Studiestargeted towards specific indoor microenvironments have alsoidentified the in-vehicle commuting microenvironment as an environ-ment in which carbon monoxide concentrations frequently exceed10 mg/m3 (9 ppm) and occasionally exceed 40 mg/m3 (35 ppm).Special environments or occurrences (indoor ice skating rinks, officeswhere emissions from parking garages migrate indoors, etc.) havebeen reported where indoor carbon monoxide levels can exceed therecommended air quality guidelines.

A majority of the targeted field studies (see section 5.6.2)monitored indoor carbon monoxide levels as a function of thepresence or absence of combustion sources (gas ranges, unvented gasand kerosene space heaters, wood-burning stoves and fireplaces andtobacco combustion). The results of these studies indicate that thepresence and use of an unvented combustion source result in indoorcarbon monoxide levels above those found outdoors. The associatedincrease in carbon monoxide concentrations can vary considerably asa function of the source, source use, condition of the source andaveraging time of the measurement. Intermittent sources such as gascooking ranges can result in high peak carbon monoxide concen-trations (in excess of 10 mg/m3 [9 ppm]), whereas long-term averageincreases in concentrations (i.e., 24-h) associated with gas ranges areconsiderably lower (on the order of 1.1 mg/m3 [1 ppm]). Thecontribution of tobacco combustion to indoor carbon monoxide levelsis variable. Under conditions of high smoking and low ventilation, thecontribution can be on the order of a few milligrams per cubic metre(parts per million). One study suggested that the contribution to


107

residential carbon monoxide concentrations of tobacco combustion ison the order of 1.1 mg/m3 (1 ppm), whereas another study showed nosignificant increase in residential carbon monoxide levels.

Unvented or poorly vented combustion sources that are used forsubstantial periods of time (e.g., unvented gas and kerosene spaceheaters) appear to be the major contributors to residential carbonmonoxide concentrations. One extensive study of unvented gas spaceheaters indicated that 12% of the homes had 15-h average carbonmonoxide concentrations greater than 9 mg/m3 (8 ppm), with thehighest concentration at 41.9 mg/m3 (36.6 ppm). Only very limiteddata are available on the contribution of kerosene heaters to theaverage carbon monoxide concentrations in residences, and these dataindicate a much lower contribution than that of gas heaters. Peakcarbon monoxide concentrations associated with both unvented gasand kerosene space heaters can exceed the current ambient 1- and 8-hstandards (40 and 10 mg/m3 [35 and 9 ppm], respectively) inresidences, and, because of the nature of the source (continuous),those peaks tend to be sustained for several hours.

Very limited data on carbon monoxide levels in residences withwood-burning stoves or fireplaces are available. Non-airtight stovescan contribute substantially to residential carbon monoxide concen-trations, whereas airtight stoves can result in small increases. Theavailable data indicate that fireplaces do not contribute measurably toaverage indoor concentrations. No information is available forsamples of residences with leaky flues. In addition, there is noinformation available on short-term indoor carbon monoxide levelsassociated with these sources, nor are there studies that examine theimpact of attached garages on residential carbon monoxideconcentrations.

The available data on short-term (1-h) and long-term (8-h)indoor carbon monoxide concentrations as a function ofmicroenvironments and sources in those microenvironments are notadequate to assess exposures in those environments. In addition, littleis known about the spatial variability of carbon monoxide indoors.These indoor microenvironments represent the most important carbonmonoxide exposures for individuals and as such need to be bettercharacterized.

Concentrations of carbon monoxide in an enclosed environmentare affected by a number of factors in addition to the source factors.


108

These factors include outdoor concentrations, proximity to outdoorsources (i.e., parking garages or traffic), volume of the space andmixing within and between indoor spaces.

Carbon monoxide measurements in enclosed spaces have beenmade either in support of total personal exposure studies or in targetedindoor studies. In the personal exposure studies, individuals wear themonitors in the course of their daily activities, taking them througha number of different microenvironments. In targeted studies, carbonmonoxide measurements are taken in indoor spaces independent ofthe activities of occupants of those spaces.

5.6.1 Indoor concentrations recorded in personal exposure studies

Three studies have reported carbon monoxide concentrations invarious microenvironments as part of an effort to measure totalhuman exposure to carbon monoxide and to assess the accuracy ofexposure estimates calculated from fixed-site monitoring data. In eachstudy, subjects wore personal carbon monoxide exposure monitors forone or more 24-h periods. Carbon monoxide concentrations wererecorded on data loggers at varying time intervals as a function oftime spent in various microenvironments. Participants kept an activitydiary in which they were asked to record time, activity (e.g., cooking),location (microenvironment type), presence and use of sources (e.g.,smokers or gas stoves) and other pertinent information. Carbonmonoxide concentrations by microenvironment were extracted fromthe measured concentrations by use of the activity diaries.

Two of the studies, conducted in Denver, Colorado, and Wash-ington, DC, by the US EPA (Hartwell et al., 1984; Johnson, 1984;Whitmore et al., 1984; Akland et al., 1985), measured the frequencydistribution of carbon monoxide exposure in a representative sampleof the urban population. The study populations were selected using amultistage sampling strategy. The third study, also conducted inWashington, DC (Nagda & Koontz, 1985), utilized a conveniencesample.

The first-mentioned Washington, DC, study obtained a total of814 person-day samples for 1161 participants, whereas the Denverstudy obtained 899 person-day samples for 485 participants. TheDenver study obtained consecutive 24-h samples for each participant,whereas the Washington, DC, study obtained one 24-h sample for


109

each participant. Both studies were conducted during the winter of1982–83.

A comparison of carbon monoxide concentrations measured inthe Washington, DC, and Denver studies is shown in Table 8 (fromAkland et al., 1985). Concentrations measured in all microenviron-ments for the Denver study were higher than those for theWashington, DC, study. This is consistent with the finding that dailymaximum 1-h and 8-h carbon monoxide concentrations at outdoorfixed monitoring sites were about a factor of 2 higher in the Denverarea than in the Washington, DC, area during the course of thestudies (Akland et al., 1985). The highest concentrations in bothstudies were associated with indoor parking garages and commuting,whereas the lowest levels were measured in indoor environmentswithout sources of carbon monoxide. Concentrations associated withcommuting are no doubt higher owing to the proximity to and densityof outside carbon monoxide sources (cars, buses and trucks),particularly during commuting hours when traffic is heaviest. Indoorlevels, especially residential levels in the absence of indoor sources,are lower primarily because of the time of day of sampling (non-commuting hours with lower outdoor levels). A more detailedbreakdown of carbon monoxide concentrations by microenvironmentsfor the Denver study is shown in Table 9 (Johnson, 1984).Microenvironments associated with motor vehicles result in thehighest concentrations, with concentrations reaching or exceeding theNAAQS 10 mg/m3 (9 ppm) reference level.

No statistical difference (P > 0.05) in carbon monoxide concen-trations was found between residences with and without gas ranges inthe Washington, DC, study. The results of a similar analysis on theDenver data, according to the presence or absence of selected indoorsources, are shown in Table 10 (Johnson, 1984); in contrast to theWashington, DC, study, the presence of an operating gas stove in theDenver study resulted in a statistically significant increase of2.97 mg/m3 (2.59 ppm). Attached garages, use of gas ranges andpresence of smokers were all shown to result in higher indoor carbonmonoxide concentrations. Concentrations were well below theNAAQS 10 mg/m3 (9 ppm) reference level but were substantiallyabove concentrations in residences without the sources.

In the second Washington, DC, study (Nagda & Koontz, 1985),197 person-days of samples were collected from 58 subjects,

Table 8. Summary of carbon monoxide exposure levels and time spent per day in selected microenvironmentsa

Microenvironment Denver, Colorado Washington, DC

n CO concentration (mean ±SE)b

Mediantime (min)

n CO concentration (mean ± SE)b Mediantime (min)

mg/m3 ppm mg/m3 ppm

Indoors, parkinggarage

31 21.5 ± 5.68 18.8 ± 4.96 14 59 11.9 ± 5.07 10.4 ± 4.43 11

In transit, car 643 9.2 ± 0.37 8.0 ± 0.32 71 592 5.7 ± 0.16 5.0 ± 0.14 79

In transit, other(bus, truck, etc.)

107 9.0 ± 0.70 7.9 ± 0.61 66 130 4.1 ± 0.34 3.6 ± 0.30 49

Outdoors, nearroadway

188 4.5 ± 0.41 3.9 ± 0.36 33 164 3.0 ± 0.23 2.6 ± 0.20 20

In transit, walking 171 4.8 ± 0.52 4.2 ± 0.45 28 226 2.7 ± 0.33 2.4 ± 0.29 32

Indoors, restaurant 205 4.8 ± 0.33 4.2 ± 0.29 58 170 2.4 ± 0.37 2.1 ± 0.32 45

Indoors, office 283 3.4 ± 0.23 3.0 ± 0.20 478 349 2.2 ± 0.31 1.9 ± 0.27 428

Indoors, store/shopping mall

243 3.4 ± 0.25 3.0 ± 0.22 50 225 2.9 ± 0.56 2.5 ± 0.49 36

Indoors, residence 776 1.9 ± 0.11 1.7 ± 0.10 975 705 1.4 ± 0.11 1.2 ± 0.10 1048

Indoors, total 776 2.4 ± 0.10 2.1 ± 0.09 1243 705 1.6 ± 0.09 1.4 ± 0.08 1332a From Akland et al. (1985).b n = number of person-days with non-zero durations; SE = standard error.

111

Table 9. Indoor microenvironments in Denver, Colorado, listed in descending order ofweighted mean carbon monoxide concentrationa

Category Number ofobservations

CO concentration (mean ± SD)b

mg/m3 ppm

Public garage 116 15.41 ± 20.77 13.46 ± 18.14

Service station or motorvehicle repair facility

125 10.50 ± 10.68 9.17 ± 9.33

Other location 427 8.47 ± 20.58 7.40 ± 17.97

Other repair shop 55 6.46 ± 8.78 5.64 ± 7.67

Shopping mall 58 5.61 ± 7.44 4.90 ± 6.50

Residential garage 66 4.98 ± 8.08 4.35 ± 7.06

Restaurant 524 4.25 ± 4.98 3.71 ± 4.35

Office 2 287 4.11 ± 4.79 3.59 ± 4.18

Auditorium, sports arena,concert hall, etc.

100 3.86 ± 5.45 3.37 ± 4.76

Store 734 3.70 ± 6.37 3.23 ± 5.56

Health care facility 351 2.54 ± 4.89 2.22 ± 4.25

Other public buildings 115 2.46 ± 3.73 2.15 ± 3.26

Manufacturing facility 42 2.34 ± 2.92 2.04 ± 2.55

Residence 21 543 2.34 ± 4.65 2.04 ± 4.06

School 426 1.88 ± 3.16 1.64 ± 2.76

Church 179 1.79 ± 3.84 1.56 ± 3.35

a Adapted from Johnson et al. (1984).b SD = standard deviation.

representing three population subgroups: housewives, office workersand construction workers. A comparison of residential carbonmonoxide concentrations from that study as a function of combustionsources and whether smoking was reported is shown in Table 11. Useof gas ranges and kerosene space heaters was found to result in higherindoor carbon monoxide concentrations. The statistical significanceof the differences was not given. Concentrations were highest inmicroenvironments associated with commuting.

5.6.2 Targeted microenvironmental studies

As demonstrated from the personal exposure studies discussedabove, individuals, in the course of their daily activities, canencounter a wide range of carbon monoxide concentrations as afunction of the microenvironments in which they spend time. Anumber of studies have been conducted over the last decade to

Table 10. Weighted means of residential exposure grouped according to the presence or absence of selected indoor carbon monoxide sources in Denver, Coloradoa

CO source CO concentration (mean ± SD)b Difference in means Significancelevel oft-testcSource present Source absent mg/m3 ppm

mg/m3 ppm mg/m3 ppm

Attachedgarage

2.62 ± 6.11 2.29 ± 5.34 2.15 ± 3.44 1.88 ± 3.00 0.47 0.41 P < 0.0005

Operatinggas stove

5.18 ± 6.98 4.52 ± 6.10 2.21 ± 4.49 1.93 ± 3.92 2.97 2.59 P < 0.0005

Smokers 3.98 ± 7.53 3.48 ± 6.58 2.16 ± 4.23 1.89 ± 3.69 1.82 1.59 P < 0.0005

a Adapted from Johnson (1984).b SD = standard deviation.c Student t-test was performed on logarithms of personal exposure monitor values.


113

Table 11. Average Washington, DC, residential carbon monoxide exposures: impactof combustion appliance use and tobacco smokinga

Appliances Reported tobacco smokingb

No Yes All cases


None 1.4 1.2 (66) 1.7 1.5 (12) 1.4 1.2 (78)

Gas stove 2.5 2.2 (15) 1.5 1.3 (1) 2.5 2.2 (16)

Kerosene spaceheater

5.8 5.1 (3) NDc 5.8 5.1 (3)

Wood burning 0.8 0.7 (2) ND 0.8 0.7 (2)

Multipleappliances

1.1 1 (1) ND 1.1 1 (1)

All cases 1.7 1.5 (87) 1.7 1.5 (13) 1.7 1.5 (100)

a Adapted from Nagda & Koontz (1985).b Percentage of subjects’ time in their own residences indicated in parentheses for each

category of appliance use and tobacco smoking.c ND = No data available.

investigate concentrations of carbon monoxide in indoormicroenvironments. These “targeted” studies have focused on indoorcarbon monoxide concentrations as a function of either themicroenvironment or sources in specific microenvironments.

5.6.2.1 Indoor microenvironmental concentrations

A number of studies have investigated carbon monoxide levelsin various indoor environments, independent of the existence ofspecific indoor sources. Major foci of these studies are microenviron-ments associated with commuting. A wide range of carbon monoxideconcentrations were recorded in these studies, with the highestconcentrations found in the indoor commuting microenvironments.These concentrations are frequently higher than concentrationsrecorded at fixed-site monitors but lower than concentrations mea-sured immediately outside the vehicles. Concentrations are generallyhigher in automobiles than in public transportation microenviron-ments. A number of the studies noted that carbon monoxide concen-trations in commuting vehicles can exceed recommended air qualityguidelines. Flachsbart et al. (1987) noted that the most important fac-tors influencing carbon monoxide concentrations inside automobiles


114

included link-to-link variability (a proxy for traffic density, vehiclemix and roadway setting), day-to-day variability (a proxy forvariations in meteorological factors and ambient carbon monoxideconcentrations) and time of day. This study noted that with increasedautomobile speed, interior carbon monoxide concentrations decreased,because grams of carbon monoxide emitted per kilometre travelleddecrease with increasing vehicle speed, and the turbulence of vehiclewake increases with increasing vehicle speed.

Service stations, car dealerships, parking garages and officespaces that have attached garages can exhibit high concentrations ofcarbon monoxide as a result of automobile exhaust. In one case(Wallace, 1983), corrective measures reduced office space carbonmonoxide concentrations originating from an attached parking garagefrom 22 mg/m3 (19 ppm) to approximately 5 mg/m3 (4 ppm). In aninvestigation of seven ice skating rinks in the Boston, Massachusetts,area, one study (Spengler et al., 1978) reported exceptionally highaverage carbon monoxide concentrations (61.4 mg/m3 [53.6 ppm]),with a high reading of 220 mg/m3 (192 ppm). Ice-cleaning machinesand poor ventilation were found to be responsible.

Residential and commercial buildings were generally found tohave low concentrations of carbon monoxide, but information isseldom provided on the presence of indoor sources or outdoor levels.

5.6.2.2 Concentrations associated with indoor sources

The major indoor sources of carbon monoxide in residences aregas ranges and unvented kerosene and gas space heaters, withproperly operating wood-burning stoves and fireplaces (non-leakyventing system) and tobacco combustion of secondary importance.Properly used gas ranges (ranges used for cooking and not spaceheating) are used intermittently and thus would contribute to short-term peak carbon monoxide levels indoors but likely would not resultin substantial increases in longer-term average concentrations.Unvented kerosene and gas space heaters typically are used for severalhours at a time and thus are likely to result in sustained higher levelsof carbon monoxide. The improper operation of gas ranges orunvented gas or kerosene space heaters (e.g., low-wick setting forkerosene heaters or yellow-tipping operation of gas ranges) couldresult in substantial increases in indoor carbon monoxide levels.Carbon monoxide levels indoors associated with tobacco combustion


115

are, based upon source emission data, expected to be low unless thereis a very high smoking density and low ventilation. In the absence ofa leaky flue or leaky fire box, indoor carbon monoxide levels fromfireplaces or stoves should be low, with short peaks associated withcharging the fire when some backdraft might occur.

The majority of studies investigating carbon monoxide concen-trations in residences, as a function of the presence or absence of aknown carbon monoxide source, typically have measured carbonmonoxide concentrations associated with the source’s use over shortperiods (on the order of a few minutes to a few hours). Only twostudies (Koontz & Nagda, 1987; Research Triangle Institute, 1990)have reported long-term average carbon monoxide concentrations(over several hours) as a function of the presence of a carbonmonoxide source for large residential sample sizes, whereas one study(McCarthy et al., 1987) reported longer-term average indoor carbonmonoxide concentrations for a small sample.

1) Average indoor-source-related concentrations

As part of a study to determine the impact of combustion sourceson indoor air quality, a sample of 382 homes in New York State, USA(172 in Onondaga County and 174 in Suffolk County), was monitoredfor carbon monoxide concentrations during the winter of 1986(Research Triangle Institute, 1990). In this study, four combustionsources were examined: gas cooking appliances, unvented kerosenespace heaters, wood-burning stoves and fireplaces and tobaccoproducts.

Gas ranges and kerosene heaters were found to result in smallincreases in average carbon monoxide levels. Use of a wood-burningstove or fireplace resulted in lower average carbon monoxide levels,presumably owing to increased air exchange rates associated with use.The study found no effect on average carbon monoxide levels withtobacco combustion and no difference by location in the residence.

Koontz & Nagda (1987), utilizing census data for sampleselection, monitored 157 homes in 16 neighbourhoods in north-central Texas, USA, over a 9-week period between January and March1985. Unvented gas space heaters were used as the primary means ofheating in 82 residences (13 had one unvented gas space heater, 36had two, and 33 had three or more) and as a secondary heat source in


116

29 residences (17 had one unvented gas space heater, and 12 had twoor more).

Residences in which unvented gas space heaters are the primaryheat source exhibited the highest carbon monoxide concentrations.Carbon monoxide concentrations were greater than or equal to10 mg/m3 (9 ppm) in 12% of the homes, with the highest concen-tration measured at 41.9 mg/m3 (36.6 ppm). No values were measuredabove 10 mg/m3 (9 ppm) for residences in which an unvented gasspace heater was not used at all or was used as a secondary heatsource. Five of the residences exceeded the 1-h, 40 mg/m3 (35 ppm)level, whereas seven of the residences exceeded the 8-h, 10 mg/m3

(9 ppm) level. Higher carbon monoxide levels were associated withmaltuned unvented gas appliances and the use of multiple unventedgas appliances.

In a study of 14 homes with one or more unvented gas spaceheaters (primary source of heat) in the area of Atlanta, Georgia, USA,McCarthy et al. (1987) measured carbon monoxide levels by con-tinuous NDIR monitors in two locations in the homes (room with theheater and a remote room in the house) and outdoors. One out of the14 unvented gas space heater homes exceeded 10 mg/m3 (9 ppm)during the sampling period. Mean indoor values ranged from 0.30 to10.87 mg/m3 (0.26 to 9.49 ppm) and varied as a function of the usepattern of the heater. Only one of the homes used more than oneheater during the air sampling. Outdoor concentrations ranged from0.34 to 1.8 mg/m3 (0.3 to 1.6 ppm).

Investigations of indoor air pollution by different heating systemsin 16 private houses in Germany (Moriske et al., 1996) showed carbonmonoxide concentrations up to 16 mg/m3 (14 ppm) (98th percentile)during the heating period and up to 4.6 mg/m3 (4.0 ppm) during thenon-heating period in 8 homes with coal burning and open fireplace.In 8 homes with central heating, carbon monoxide concentrationswere up to 2.3 mg/m3 (2.0 ppm) during both heating and non-heatingperiods. In a home on the ground floor of a block of flats with acentral stove in the basement below, carbon monoxide concentrationswere up to 64 mg/m3 (56 ppm) during the heating season.


117

2) Peak indoor-source-related concentrations

Short-term or peak indoor carbon monoxide concentrationsassociated with specific sources were obtained for a few field studies.In these studies, a wide range of peak carbon monoxideconcentrations was observed in various residences with differentindoor carbon monoxide sources. The highest concentrationsmeasured (>700 mg/m3 [>600 ppm]) were associated with emissionsfrom geysers (water heaters), found in a large study conducted in theNetherlands (Brunekreef et al., 1982). Peak levels of carbon monoxideassociated with gas ranges were from 1.1 mg/m3 (1.0 ppm) to morethan 110 mg/m3 (100 ppm). This broad range is somewhat consistentwith the results of other studies evaluating carbon monoxideemissions from gas ranges. The variability is in part due to thenumber of burners used, flame condition, condition of the burners,etc. As might be expected, radiant kerosene heaters produced highercarbon monoxide concentrations than did convective heaters.Unvented gas space heaters were generally associated with highercarbon monoxide peaks than were gas ranges or kerosene heaters. Asnoted above, the peaks associated with gas or kerosene heaters arelikely to be sustained over longer periods of time because of the longsource-use times.

Test houses have been used by investigators to evaluate theimpact of specific sources, modifications to sources and variations intheir use on residential peak carbon monoxide concentrations.

In one of the earliest investigations of indoor air quality, Wadeet al. (1975) measured indoor and outdoor carbon monoxide levels infour houses that had gas stoves. Indoor concentrations were found tobe 1.7–3.8 times higher than the outdoor levels. Carbon monoxidelevels in one house exceeded 10 mg/m3 (9 ppm), the NAAQSreference level. As part of a modelling study of emissions from a gasrange, Davidson et al. (1987) measured carbon monoxide concentra-tions in three residences. Peak carbon monoxide levels in excess of6 mg/m3 (5 ppm) were measured in one townhouse.

Indoor carbon monoxide levels associated with wood-burningstoves were measured in two test house studies. In one study(Humphreys et al., 1986), indoor carbon monoxide levels associatedwith the use of both airtight (conventional and catalytic) and non-airtight wood heaters were evaluated in a 337-m3 weatherized home.Indoor carbon monoxide concentrations were higher than outdoor


118

levels for all tests. Conventional airtight stoves produced indoorcarbon monoxide levels typically about 1.1–2.3 mg/m3 (1–2 ppm)above background level, with a peak concentration of 10.4 mg/m3

(9.1 ppm). Use of non-airtight stoves resulted in average indoorcarbon monoxide concentrations 2.3–3.4 mg/m3 (2–3 ppm) aboveoutdoor concentrations, with peak concentrations as high as33.9 mg/m3 (29.6 ppm). In a 236-m3 house (Traynor et al., 1984), fourwood-burning stoves (three airtight and one non-airtight) were tested.The airtight stoves generally resulted in small contributions to bothaverage and peak indoor carbon monoxide levels (0.11–1.1 mg/m3

[0.1–1 ppm] for the average and 0.23–3.1 mg/m3 [0.2–2.7 ppm] forthe peak). The non-airtight stove contributed as much as 10.4 mg/m3

(9.1 ppm) to the average indoor level and 49 mg/m3 (43 ppm) to thepeak.

3) Indoor concentrations related to environmental tobacco smoke

Carbon monoxide has been measured extensively in chamberstudies as a surrogate for environmental tobacco smoke (e.g., Bridge& Corn, 1972; Hoegg, 1972; Weber et al., 1976, 1979a,b; Leadereret al., 1984; Weber, 1984; Clausen et al., 1985). Under steady-stateconditions in chamber studies, where outdoor carbon monoxide levelsare monitored and the tobacco brands and smoking rates are con-trolled, carbon monoxide can be a reasonably good indicator ofenvironmental tobacco smoke and is used as such. Under such cham-ber conditions, carbon monoxide concentrations typically range fromless than 1.1 mg/m3 (1 ppm) to greater than 11 mg/m3 (10 ppm).

A number of field studies have monitored carbon monoxide indifferent indoor environments with and without smoking occupants.Although carbon monoxide concentrations were generally higher inindoor spaces when smoking occurred, the concentrations were highlyvariable. The variability of carbon monoxide production from tobaccocombustion, the variability in the number of cigarettes smoked anddifferences in ventilation and variability of outdoor concentrationsmake it difficult to assess the contribution of tobacco combustion toindoor carbon monoxide concentrations. The chamber studies andfield studies conducted do indicate that under typical smoking condi-tions encountered in residences or offices, carbon monoxide concen-trations can be expected to be above background outdoor levels, butlower than the levels resulting from other unvented combustionsources. In indoor spaces where heavy smoking occurs and in small


119

indoor spaces, carbon monoxide emissions from tobacco combustionwill be an important contributor to carbon monoxide concentrations.

5.7 Occupational exposure

Carbon monoxide is a ubiquitous contaminant occurring in avariety of occupational settings. The number of persons occupationallyexposed to carbon monoxide in the working environment is greaterthan for any other physical or chemical agent (Hosey, 1970), withestimates as high as 975 000 occupationally exposed at high levels inthe USA (NIOSH, 1972).

Two main sources for background exposures in both occupationaland non-occupational settings appear to be smoking and the internalcombustion engine (NAS, 1969). Also, endogenous carbon monoxidemay be derived from certain halomethanes that are biotransformed bymixed-function oxidases in vivo. Smoking is a personal habit thatmust be considered in evaluating exposure in general, as well as thoseoccurring in workplaces. In addition, work environments are oftenlocated in densely populated areas, and such areas frequently have ahigher background concentration of carbon monoxide compared withless densely populated residential areas. Thus, background exposuresmay be greater during work hours than during non-work hours.

There are several sources other than smoking and the internalcombustion engine that contribute to exposure during work hours.These include contributions to background levels by combustion oforganic materials in the geographic area of the workplace; work inspecific industrial processes that produce carbon monoxide; and expo-sure to halomethanes that give rise to endogenous carbon monoxideduring biotransformation. Therefore, a number of potential sources ofcarbon monoxide should be considered when evaluating the riskassociated with carbon monoxide exposure, including personal habits,living conditions and co-exposure to other potential sources of carbonmonoxide or to xenobiotics that are metabolized to carbon monoxideor that interfere with biotransformation processes. Finally, theparticular vulnerability of specific groups at increased risk because ofsome physiological (pregnancy) or pathological (angina, anaemia,respiratory insufficiency) conditions should be taken into account inthe health surveillance of occupationally exposed workers.


120

5.7.1 Occupational exposure limits

Elevated concentrations of carbon monoxide occur in numeroussettings, including those at work, at home or in the street. Acuteeffects related to production of anoxia from exposures to carbonmonoxide have historically been a basis for concern. In recent years,however, this concern has grown to include concerns for potentialeffects from chronic exposure as well (Sammons & Coleman, 1974;Rosenstock & Cullen, 1986a,b).

Three kinds of occupational exposure limits are currently utilizedin a number of countries (Table 12). The most common are averagepermissible concentrations for a typical 8-h working day (time-weighted average, TWA), concentrations for short-term exposures,generally of 15-min duration (short-term exposure limit, STEL), andmaximum permissible concentrations not to be exceeded (ceilinglimit).

Whereas some countries do not apply legally binding occupa-tional exposure limits and others refer to maximum allowable concen-trations, threshold limit values (TLVs) are the most widely used andaccepted standards representing “conditions under which it is believedthat nearly all workers may be repeatedly exposed day after daywithout adverse health effects.” In the introduction to the list of TLVspublished by the American Conference of Governmental IndustrialHygienists (ACGIH, 1995), it is also stressed that these limits are notfine lines separating safe from dangerous situations, but ratherguidelines to be used by professionals with a specific training inindustrial hygiene. Moreover, it is recognized that a small percentageof workers may be “unusually responsive to some industrial chemicalsbecause of genetic factors, age, personal habits (smoking, alcohol, orother drugs), medication, or previous exposure. Such workers may notbe adequately protected from adverse health effects at concentrationsat or below the threshold limit. An occupational physician shouldevaluate the extent to which such workers require additionalprotection” (ACGIH, 1995).

A TWA concentration of 29 mg/m3 (25 ppm) has been recom-mended by ACGIH as the TLV for carbon monoxide since 1991. Such


121

Table 12. Worldwide occupational exposure limits for carbon monoxidea

TWA STEL Ceiling


Austria 33 30 – – – –

Belgium 55 50 – – – –

Brazil 43 39 – – – –

Bulgaria 20 – – – – –

Chile 44 40 – – – –

China, P.R. of 30 – – – – –

China (Taiwan) 55 50 – – – –

Czechoslovakia 30 – – – 150 –

Denmark 40 35 – – – –

Egypt – 100 – – – –

Finland 55 50 85 75 – –

France – 50 – – – –

FRG 33 30 – – – –

Holland 55 50 – – – –

Hungary 20 – 100 – – –

India 55 50 440 400 – –

Indonesia 115 100 – – – –

Italy 55 50 – – – –

Japan 55 50 – – – –

Mexico 55 50 – – – –

Poland 20 – – – – –

Romania 30 – 50 – – –

Sweden 40 35 120 100 – –

Switzerland 33 30 – – – –

United Kingdom 55 50 440 400 – –

USA – 25 – 400 – 200

USSR 33 30 – – 20 –

Venezuela 55 50 – – 440 400

Yugoslavia 58 50 – – – –

a From ACGIH (1987); CEC (1993).


122

a limit is likely to be adopted by other countries traditionally relyingon ACGIH-recommended TLVs and by countries relying on therecommended exposure limits (RELs) adopted by the US NationalInstitute for Occupational Safety and Health (NIOSH), currently setat 40 mg/m3 (35 ppm). Although ceiling limits and STELs are notexpressly indicated, general rules advise that exposure levels “mayexceed the TLV-TWA three times for no more than 30 min during awork-day, and under no circumstances should they exceed five timesthe TLV-TWA.” Therefore, a ceiling of 143 mg/m3 (125 ppm) and aSTEL of 86 mg/m3 (75 ppm) may be derived from such a guidance.Biotransformation into carbon monoxide rather than other toxicproperties is the scientific basis for setting the TLV for methylenechloride at 177 mg/m3 (50 ppm).

In addition to ambient monitoring, biomarkers can be used toassess exposure, susceptibility and early effects. Carboxyhaemoglobinis a widely accepted biomarker — or biological exposure index (BEI),according to ACGIH nomenclature — of exposure to carbonmonoxide. The ACGIH has recently proposed a carboxyhaemoglobinlevel of 3.5% as a BEI being most likely reached by a non-smoker atthe end of an 8-h exposure to 29 mg carbon monoxide/m3 (25 ppm)(ACGIH, 1991). The biological tolerance limit (BAT) recommendedby the DFG (1996) is 5% carboxyhaemoglobin, corresponding to acarbon monoxide TWA level of 34 mg/m3 (30 ppm).

Neither the BEI nor the BAT is applicable to tobacco smokers.According to Gilli et al. (1979), a serum thiocyanate level above 3.8mg/litre and a carboxyhaemoglobin level in the range of 2.5–6% area certain indication of cigarette smoking, whereas a thiocyanateconcentration below 3.8 mg/litre and a carboxyhaemoglobin levelabove 5% indicate occupational exposure to carbon monoxide.Tobacco smoking can also be assessed measuring cotinine excretion,but studies relating cotinine in urine to carboxyhaemoglobin are notavailable. An alternative BEI is carbon monoxide concentration inend-expired air collected at the end of the working shift. A concen-tration of 23 mg/m3 (20 ppm) would be reached after exposure to an8-h TWA level of 29 mg/m3 (25 ppm).

It ought to be noted that both BEIs and BATs are usuallyestablished taking into account the relationship between biomarkersand exposure levels rather than the relationship between biomarkersand adverse effects. Therefore, higher carboxyhaemoglobin levels aregenerally accepted for smokers, taking into account the fact that theirhigher levels result from a voluntary habit giving rise to an additional


123

carbon monoxide burden. However, if adverse effects are expected tooccur as a consequence of carboxyhaemoglobin and carbon monoxidebody burden, then smokers should be advised about their additionalrisk.

NIOSH (1972) observed that “the potential for exposure tocarbon monoxide for employees in the work place is greater than forany other chemical or physical agent” and recommended thatexposure to carbon monoxide be limited to a concentration no greaterthan 40 mg/m3 (35 ppm), expressed as a TWA for a normal 8-hworkday, 5 days per week. A ceiling concentration was alsorecommended at a limit of 230 mg/m3 (200 ppm), not to exceed anexposure time greater than 30 min. Occupational exposures at theproposed concentrations and conditions underlying the basis of thestandard were considered to maintain carboxyhaemoglobin in bloodbelow 5%.

Although it was not stated, the basis of the recommended NIOSHstandard (i.e., maintaining carboxyhaemoglobin below 5% in blood)assumes that (1) the sensitive population group protected by airquality standards would not be occupationally exposed and (2)contributions from other non-occupational sources would also be lessthan a TWA concentration of 40 mg/m3 (35 ppm). It was recognizedthat such a standard may not provide the same degree of protection tosmokers, for example. Other particularly vulnerable groups of workersdeserve special consideration. Among these, pregnant women shouldrequire special protection because of the potentially deleterious effectsof carbon monoxide exposure on the fetuses (NIOSH, 1972). Thesame requirement for safety of pregnant women has been establishedby the DFG (1996).

Although recognizing that biological changes might occur at thelow level of exposure recommended in the proposed standard, NIOSHconcluded that subtle aberrations in the nervous system withexposures producing carboxyhaemoglobin concentrations in blood ator below 5% did not demonstrate significant impairments that wouldcause concern for the health and safety of workers. In addition,NIOSH observed that individuals with impairments that interfere withnormal oxygen delivery to tissues (e.g., emphysema, anaemia,coronary heart disease) may not have the same degree of protection ashave less impaired individuals. It was also recognized that work athigher altitudes (e.g., 1500–2400 m above sea level) would necessitate


124

decreasing the exposure limit below 40 mg/m3 (35 ppm), tocompensate for a decrease in the oxygen partial pressure as a result ofhigh-altitude environments and a corresponding decrease inoxygenation of the blood (NIOSH, 1972). High-altitude environmentsof concern include airline cabins at a pressure altitude of 1500 m orgreater (NRC, 1986b) and high mountain tunnels (Miranda et al.,1967).

5.7.2 Exposure sources

The contribution of occupational exposures to carbon monoxidecan be separated from other sources of carbon monoxide exposure, butthere are at least two conditions to consider:

(1) When carbon monoxide concentrations at work are higherthan the carbon monoxide equilibrium concentration associ-ated with the percent carboxyhaemoglobin at the start of thework shift, there will be a net absorption of carbon monox-ide and an increase in percent carboxyhaemoglobin. Non-smokers will show an increase that is greater than that forsmokers because they start from a lower baseline carboxy-haemoglobin level. In some cases, non-smokers may showan increase and smokers a decrease in percent carboxy-haemoglobin.

(2) When carbon monoxide concentrations at work are lowerthan the equilibrium concentration necessary to produce theworker’s current level of carboxyhaemoglobin, then thepercent carboxyhaemoglobin will show a decrease. Therewill be a net loss of carbon monoxide at work.

As mentioned above, occupational exposures can stem from threesources: (1) through background concentrations of carbon monoxide,(2) through work in industrial processes that produce carbonmonoxide as a product or by-product and (3) through exposure tosome halomethanes that are metabolized to carbon monoxide in vivo.In addition, work environments that tend to accumulate carbonmonoxide concentrations may result in occupational exposures.Rosenman (1984) lists a number of occupations in which the workersmay be exposed to high carbon monoxide concentrations. This listincludes acetylene workers, blast furnace workers, coke oven workers,diesel engine operators, garage mechanics, steel workers, metal oxide


125

reducers, miners, nickel refining workers, organic chemical synthesiz-ers, petroleum refinery workers and pulp and paper workers. Inaddition, because methylene chloride is metabolized to carbonmonoxide in the body, aerosol packagers, anaesthetic makers, bitumenmakers, degreasers, fat extractors, flavouring makers, leather finishworkers, oil processors, paint remover makers, resin makers andworkers exposed to dichloromethane-containing solvent mixtures andstain removers can also have high carboxyhaemoglobin levels.

5.7.3 Combined exposure to xenobiotics metabolized to carbonmonoxide

Certain halomethanes, particularly dichloromethane, a frequentlyused organic solvent also known as methylene chloride (reviewed inEHC 164), are metabolized to carbon monoxide, carbon dioxide andchlorine (or iodine or bromine) in a reaction catalysed by cytochromeP-450 2E1.

Combined exposures frequently occur at the workplace, and theirtemporal sequence may result in opposite effects. Concurrent exposureto other substrates of cytochrome P-450 2E1 — including ethanol anda number of common organic solvents, such as benzene and its alkylderivatives, trichloroethylene and acetone — may cause a competitiveinhibition of the oxidation of methylene chloride to carbon monoxide.As the same substances may act as inducers, provided they are nolonger present to compete with methylene chloride, increased rates ofcarbon monoxide formation are expected to occur in peoplechronically exposed to such solvents or to ethanol and then exposedto methylene chloride at the workplace. As a corollary of theseobservations, and taking into account the fact that the carbonmonoxide produced from inhaled dichloromethane rather than theparent compound underlies non-cancer end-points and the fact thatthe carbon monoxide formation rate is modified in a different way byprior and concurrent exposure to other substances, it would seemlogical to rely on the same BEIs proposed for carbon monoxide tomonitor exposure to methylene chloride and other halomethanesrather than on ambient monitoring alone.

5.7.4 Typical studies at the workplace

Aircraft accidents involving 113 aircraft, 184 crew members and207 passengers were investigated to characterize accident toxicology


126

and to aid in the search for causation of a crash (Blackmore, 1974).Determinations of percent carboxyhaemoglobin in blood samplesobtained from victims enabled differentiation of a variety of accidentsequences involving fires. For example, percent carboxyhaemoglobindeterminations combined with passenger seating information andcrew assignments can help differentiate between fire in flight and fireafter the crash, between survivability of the crash and death due tosmoke inhalation, and between specific malfunctions in equipmentoperated by a particular crew member and defects in space heating inthe crew cabin or passenger compartment. One accident in the serieswas associated with a defective space heater in the crew compartment.Another accident was also suspicious with regard to a space heater.

Carbon monoxide concentrations were used to classify workersfrom 20 foundries into three groups: those with definite occupationalexposure, those with slight exposure and controls (Hernberg et al.,1976). Angina pectoris, electrocardiogram findings and blood pres-sures of foundry workers were evaluated in terms of carbon monoxideexposure for the 1000 workers who had the longest occupationalexposures for the 20 foundries. Angina showed a clear dose–responsewith exposure to carbon monoxide either from occupational sourcesor from smoking, but there was no such trend in electrocardiogramfindings. The systolic and diastolic pressures of carbon monoxide-exposed workers were higher than those for other workers, when ageand smoking habits were considered.

Carboxyhaemoglobin and smoking habits were studied for apopulation of steelworkers and compared with those for blast furnaceworkers as well as employees not exposed at work (Jones & Walters,1962). Carbon monoxide is produced in coke ovens, blast furnacesand sintering operations. Exhaust gases from these operations areoften used for heating and as fuels for other processes. Fifty-sevenvolunteers working in the blast furnace area were studied for smokinghabits, symptoms of carbon monoxide exposure and estimations ofcarboxyhaemoglobin levels by an expired air technique. The mainincrease in carboxyhaemoglobin for blast furnace personnel was 2.0%for both smokers and non-smokers in the group. For smokers in theunexposed control group, there was a decrease in percent carboxy-haemoglobin. A follow-up study found similar results (Butt et al.,1974). Virtamo & Tossavainen (1976) reported a study of carbonmonoxide measurements in air of 67 iron, steel or copper alloy


127

foundries. Blood carboxyhaemoglobin of iron workers exceeded 6%in 26% of the non-smokers and 71% of the smokers studied.

Poulton (1987) found that a medical helicopter with its enginerunning in a narrowed or enclosed helipad was a source of potentialexposure to carbon monoxide, JP-4 fuel and possibly other combustionproducts for flight crews, medical personnel, bystanders and patientsbeing evacuated. Measurements were made by means of a portableinfrared analyser. Carbon monoxide concentrations were found to begreatest near the heated exhaust. Concentrations ranged from 9 to 49mg/m3 (8 to 43 ppm).

Exhaust from seven of the most commonly used chain-saws(Nilsson et al., 1987) was analysed under laboratory conditions tocharacterize emissions. The investigators conducted field studies onexposures of loggers using chain-saws in felling operations and alsoin limbing and bucking into lengths. In response to an inquiry, 34%of the loggers responded that they often experienced discomfort fromthe exhaust fumes of chain-saws, and another 50% complained ofoccasional problems. Sampling for carbon monoxide exposures wascarried out for 5 days during a 2-week work period in a sparse pinestand at an average wind speed of 0–3 m/s, a temperature range of1–16 °C and a snow depth of 50–90 cm. Carbon monoxide concen-trations ranged from 10 to 23 mg/m3 (9 to 20 ppm), with a mean valueof 20.0 mg/m3 (17.5 ppm). Carbon monoxide concentrationsmeasured under similar, but snow-free, conditions ranged from 24 to44 mg/m3 (21 to 38 ppm), with a mean value of 32 mg/m3 (28 ppm).In another study, carbon monoxide exposures were monitored for non-smoking chain-saw operators; average exposures recorded were from23 to 63 mg/m3 (20 to 55 ppm), with carboxyhaemoglobin levelsranging from 1.5 to 3.0% (Van Netten et al., 1987).

Forklift operators, stevedores and winch operators were moni-tored for carbon monoxide in expired air, using a Mine SafetyAppliances analyser, to calculate percent carboxyhaemoglobin(Breysse & Bovee, 1969). Periodic blood samples were collected tovalidate the calculations. Forklift operators and stevedores, but notwinch operators, work in the holds of ships. The ships to be evaluatedwere selected on the basis of their use of gasoline-powered forklifts foroperations. To evaluate seasonal variations in percent carboxy-haemoglobin, analyses were performed for one 5-day period per


128

month for a full year. Efforts were made to select a variety of ships forevaluation. In total, 689 determinations of percent carboxyhaemo-globin were made from samples of blood to compare with values fromsamples of expired air. The samples were collected on 51 separatedays involving 26 different ships. Smoking was found to be a majorcontributing factor to the percent carboxyhaemoglobin levels found.Carboxyhaemoglobin values for non-smokers indicated that the useof gasoline-powered lifts in the holds of the ships did not produce acarbon monoxide concentration in excess of 57 mg/m3 (50 ppm) forup to 8 h as a TWA under the work rules and operating conditions inpractice during the study. Smoking behaviour confounded exposureevaluations.

Carbon monoxide concentrations have been measured in avariety of workplaces where potential exists for accumulation fromoutside sources. Exposure conditions in workplaces, however, aresubstantially different. The methods to be applied, groupcharacteristics, jobs being performed, smoking habits and physicalcharacteristics of the facilities themselves introduce considerablevariety in the approaches used. Typical studies are discussed below.

Wallace (1983) investigated carbon monoxide in air and breathof employees working at various times over a 1-month period in anoffice constructed in an underground parking garage. Carbon monox-ide levels were determined by use of a device containing a proprietarysolid polymer electrolyte to detect electrons emitted in the oxidationof carbon monoxide to carbon dioxide. Variation in carbon monoxidemeasurements in ambient air showed a strong correlation with trafficactivity in the parking garage. Initially, the office carbon monoxidelevels were found to be at a daily average of 21 mg/m3 (18 ppm), withthe average from 12:00 to 4:00 p.m. at 25 mg/m3 (22 ppm) and theaverage from 4:00 to 5:00 p.m. at 41 mg/m3 (36 ppm). Analyses ofexpired air collected from a group of 20 non-smokers working in theoffice showed a strong correlation with concentrations of carbonmonoxide in ambient air and traffic activity. For example, the averagecarbon monoxide in expired air for one series of measurements was26.8 mg/m3 (23.4 ppm), compared with simultaneous measurementsof carbon monoxide concentrations in air of 25–30 mg/m3

(22–26 ppm). After a weekend, carbon monoxide concentrations inbreath on Monday morning were substantially decreased (around8 mg/m3 [7 ppm]), but they rose again on Monday afternoon to equal


129

the air levels of 14 mg/m3 (12 ppm). Closing fire doors and usingexisting garage fans decreased carbon monoxide concentrations in thegarage offices to 2.3 mg/m3 (2 ppm) or less, concentrations similar tothose for other offices in the complex that were located away from thegarage area.

Carboxyhaemoglobin levels (Ramsey, 1967) were determinedover a 3-month period during the winter for 38 parking garage atten-dants, and the values for carboxyhaemoglobin were compared withvalues from a group of 27 control subjects. Blood samples werecollected by finger stick on Monday mornings at the start of the workweek, at the end of the work shift on Mondays and at the end of thework week on Friday afternoons. Hourly analyses were carried out onthree different weekdays using potassium paladosulfite indicator tubesfor the concentrations of carbon monoxide at three of the six garagesin the study. Hourly levels ranged from 8 to 270 mg/m3 (7 to240 ppm), and the composite mean of the 18 daily averages was67.4 ± 28.5 mg/m3 (58.9 ± 24.9 ppm). Although the Monday versusFriday afternoon values for carboxyhaemoglobin were not signifi-cantly different, there were significant differences between Mondaymorning and Monday afternoon values. Smokers showed higherstarting baseline levels, but there was no apparent difference in netincrease in carboxyhaemoglobin body burden between smokers andnon-smokers. Carboxyhaemoglobin levels for non-smokers rangedfrom a mean of 1.5 ± 0.83% for the morning samples to 7.3 ± 3.46%for the afternoon samples. For smokers, these values were2.9 ± 1.88% for the morning and 9.3 ± 3.16% for the afternoon. Theauthors observed a crude correlation between daily average carbonmonoxide levels in air and carboxyhaemoglobin levels observed fora 2-day sampling period.

In a study of motor vehicle examiners conducted by NIOSH(Stern et al., 1981), carbon monoxide levels of 5–24 mg/m3

(4–21 ppm) TWA were recorded in six outdoor motor vehicleinspection stations. In contrast, the semi-open and enclosed stationshad levels of 11–46 mg/m3 (10–40 ppm) TWA. The levels exceededthe recommended NIOSH standard of 40 mg/m3 (35 ppm) TWA on10% of the days sampled. In addition, all stations experienced peakshort-term levels above 230 mg/m3 (200 ppm).


130

Carboxyhaemoglobin levels were measured for 22 employees ofan automobile dealership during the winter months when garagedoors were closed and ceiling exhaust fans were turned off (Andrecset al., 1979). Employees subjected to testing included garagemechanics, secretaries and sales personnel. These included 17 malesaged 21–37 and five females aged 19–36. Blood samples werecollected on a Monday morning before the start of work and on Fridayat the end of the work week. Smokers working in the garage areashowed a Monday morning mean carboxyhaemoglobin value of4.87 ± 3.64% and a Friday afternoon mean value of 12.9 ± 0.83%.Non-smokers in the garage showed a corresponding increase incarboxyhaemoglobin, with a Monday morning mean value of1.50 ± 1.37% and a Friday afternoon mean value of 8.71 ± 2.95%.Non-smokers working in areas other than the garage had a Fridayafternoon mean value of 2.38 ± 2.32%, which was significantly lowerthan the mean values for smokers and non-smokers in the garagearea. Environmental concentrations or breathing zone samples forcarbon monoxide were not collected. The authors concluded thatsmokers have a higher baseline level of carboxyhaemoglobin than donon-smokers, but both groups show similar increases in carboxy-haemoglobin during the work week while working in the garage area.The authors observed that the concentrations of carboxyhaemoglobinfound in garage workers were the same as those reported to produceneurological impairment. These results are consistent with thosereported by Amendola & Hanes (1984), who reported some of thehighest indoor levels collected at automobile service stations anddealerships. Concentrations ranged from 18.5–126.9 mg/m3

(16.2–110.8 ppm) in cold weather to 2.5–24.7 mg/m3 (2.2–21.6 ppm)in warm weather.

A group of 34 employees, 30 men and 4 women, working inmultistorey garages was evaluated for exposures to exhaust fumes(Fristedt & Akesson, 1971). Thirteen were service employees workingat street level, and 21 were shop employees working either one storeyabove or one storey below street level. Six facilities were included inthe study. Blood samples were collected on a Friday at four facilities,on Thursday and Friday at another and on a Thursday only at a sixthfacility. The blood samples were evaluated for red blood cell andwhite blood cell counts, carboxyhaemoglobin, lead and *-amino-levulinic acid. Work histories, medical case histories and smokinghabits were recorded. Among the employees evaluated, 11 of24 smokers and 3 of 10 non-smokers complained of discomfort from


131

exhaust fumes. Smokers complaining of discomfort averaged 6.6%carboxyhaemoglobin, and non-smokers complaining averaged 2.2%carboxyhaemoglobin. The corresponding values for non-complainingworkers averaged 4.2 and 1.1%, respectively.

Air pollution by carbon monoxide in underground garages wasinvestigated as part of a larger study of traffic pollutants in Paris,France (Chovin, 1967). Work conducted between the hours of8:00 a.m. and 10:00 p.m. resulted in exposures in excess of 57 mg/m3

(50 ppm) and up to 86 mg/m3 (75 ppm), on a TWA basis.

As part of a larger study of carbon monoxide concentrations andtraffic patterns in Paris (Chovin, 1967), samples were taken in roadtunnels. There was good correlation between the traffic volumescombined with the lengths of the tunnels and the carbon monoxideconcentrations found. None of the tunnels studied had mechanicalventilation. The average carbon monoxide concentrations in thetunnels were 31 mg/m3 (27 ppm) and 34 mg/m3 (30 ppm) for 1965and 1966, respectively, compared with an average of 27 mg/m3

(24 ppm) in the streets for both years. The average risk for a personworking or walking in a street or tunnel was considered by the authorsto be 3–4 times less than the maximal risk indicated by values forcarbon monoxide from instantaneous air sample measurements. In theUSA, Evans et al. (1988) studied bridge and tunnel workers inmetropolitan New York City, New York. The average carboxyhaemo-globin concentration over the 11 years of study averaged 1.73% fornon-smoking bridge workers and 1.96% for tunnel workers.

In a discussion of factors to consider in carbon monoxide controlof high-altitude highway tunnels, Miranda et al. (1967) reviewed thehistories of several tunnels. Motor vehicles were estimated to emitabout 0.03 kg carbon monoxide/km at sea level. At 3350 m and agrade of 1.64%, emissions were estimated at 0.1 kg/km (for vehiclesmoving upgrade). Tunnels with ventilation are generally designed tocontrol carbon monoxide concentrations at or below 110 mg/m3

(100 ppm). The Holland Tunnel in New York was reported to average74 mg/m3 (65 ppm), with a recorded maximum of 418 mg/m3

(365 ppm) due to a fire. For the Sumner Tunnel in Boston, Massa-chusetts, ventilation is started at carbon monoxide concentrations of110 mg/m3 (100 ppm), and additional fans are turned on and an alarmis sounded at 290 mg/m3 (250 ppm). The average carbon monoxide


132

concentration is 57 mg/m3 (50 ppm). The Mont Blanc Tunnel is11.6 km long at an average elevation of 1274 m. This tunnel isdesigned to maintain carbon monoxide concentrations at or below110 mg/m3 (100 ppm). The Grand Saint Bernard Tunnel is 5.6 kmlong at an average elevation of 1830 m. The tunnel is designed tomaintain carbon monoxide concentrations at or below 230 mg/m3

(200 ppm). For the tunnel at 3350 m, the authors recommendedmaintaining carbon monoxide concentrations at or below 29 mg/m3

(25 ppm) for long-term exposures and at or below 57 mg/m3 (50 ppm)for peaks of 1-h exposure. The recommendations are based onconsiderations of a combination of hypoxia from lack of oxygen dueto the altitude and stress of carbon monoxide exposures of workersand motorists.

Carbon monoxide exposures of tollbooth operators were studiedalong the New Jersey, USA, Turnpike. The results reported byHeinold et al. (1987) indicated that peak exposures for 1 h rangedfrom 14 to 27 mg/m3 (12 to 24 ppm), with peak 8-h exposures of 7–17mg/m3 (6–15 ppm).

Carboxyhaemoglobin levels were determined for 15 non-smokersat the start, middle and end of a 40-day submarine patrol (Bondiet al., 1978). Values found were 2.1%, 1.7% and 1.7%, respectively.The average carbon monoxide concentration in ambient air was 8mg/m3 (7 ppm). The authors observed that the levels of percentcarboxyhaemoglobin found would not cause significant impairmentof the submariners.

In contrast, Iglewicz et al. (1984) found in a 1981 study thatcarbon monoxide concentrations inside ambulances in New Jerseywere often above the US EPA 8-h standard of 10 mg/m3 (9 ppm). Forexample, measurements made at the head of the stretcher exceeded10 mg/m3 (9 ppm) on nearly 27% of the 690 vehicles tested, with4.2% (29 vehicles) exceeding 40 mg/m3 (35 ppm).

Environmental tobacco smoke has been reviewed (NRC, 1986b)for contributions to air contaminants in airliner cabins and topotential exposures for passengers and flight crew members.Environmental tobacco smoke is described as a complex mixturecontaining many components. Analyses of carbon monoxide contentand particulate matter in cabin air were used as surrogates for the


133

vapour phases and solid components of environmental tobacco smoke,respectively. A mathematical model was developed and used tocalculate the dilution of contaminants by outside make-up air. Theamount of carbon monoxide in the cabin environment depends on therate and number of cigarettes smoked and on the rate of dilution byoutside make-up air. An additional factor to consider is the influenceof pressure altitude on the absorption of carbon monoxide and othergases. The legal limit for pressure altitude is 2440 m. The partialpressure of oxygen is 16 kPa assuming 20% oxygen in the cabin air,compared with 20 kPa at sea level. It is possible that the absorptionrate for carbon monoxide would be increased under hypobaricconditions.

A study of municipal bus drivers in the San Francisco Bay,California, USA, area by Quinlan et al. (1985) showed a TWA of1.1–26 mg/m3 (1–23 ppm), with a mean TWA of 6.3 mg/m3 (5.5 ppm)and standard deviation of 5.6 mg/m3 (4.9 ppm). The peak exposuresranged from 8 to 54 mg/m3 (7 to 47 ppm), with a mean of 29.0 mg/m3

(25.3 ppm) and standard deviation of 14.3 mg/m3 (12.5 ppm).

Cooke (1986) reported finding no significant increases outsidenormal ranges, compared with the general population, for levels ofblood lead and carboxyhaemoglobin in a group of 13 roadsideworkers. Samples were collected in the afternoon of a workday.Among the subjects, 7 of 13 were smokers and showed percentcarboxyhaemoglobin in blood ranging from 3.0 to 8.8% (mean of5.5%). For non-smokers, percent carboxyhaemoglobin ranged from0.5 to 1.4% (mean of 1.2%). Each smoker had smoked at least onecigarette in the 4 h preceding collection of blood samples. No sampleswere collected before the start of work, and no measurements ofcarbon monoxide in air at the work sites were presented.

2. chemistry and analytical methods 2.1 … · 2. chemistry and analytical methods ... chemistry...

Documents